<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Extra Credit-A Tech Blog by Guild - Medium]]></title>
        <description><![CDATA[Welcome to Extra Credit, Guild’s technology blog! We are working to make education affordable and achievable for 60+ million working adults in the US and we are sharing our technology learnings along the way. - Medium]]></description>
        <link>https://medium.com/extra-credit-by-guild?source=rss----50d55bb8b267---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Extra Credit-A Tech Blog by Guild - Medium</title>
            <link>https://medium.com/extra-credit-by-guild?source=rss----50d55bb8b267---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 08:42:03 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/extra-credit-by-guild" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Eligibility at Guild — Adventures in Event Processing on AWS]]></title>
            <link>https://medium.com/extra-credit-by-guild/eligibility-at-guild-adventures-in-event-processing-on-aws-8ebea36256f7?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/8ebea36256f7</guid>
            <category><![CDATA[edtech]]></category>
            <category><![CDATA[systems-engineering]]></category>
            <category><![CDATA[serverless]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[guild-education]]></category>
            <dc:creator><![CDATA[chap]]></dc:creator>
            <pubDate>Thu, 15 Oct 2020 18:01:05 GMT</pubDate>
            <atom:updated>2020-10-15T18:01:03.153Z</atom:updated>
            <content:encoded><![CDATA[<h3>Eligibility at Guild — Adventures in Event Processing on AWS</h3><h3>Introduction</h3><p><a href="https://www.guildeducation.com/">Guild Education</a> is a marketplace that partners with employers to provide education benefits to their employees. Guild’s value proposition derives from our domain expertise and the infrastructure that we’ve built around administering these benefits. Operationally, some of the most important questions we strive to answer effectively and accurately are:</p><blockquote>What is available to employee X as part of their benefit?</blockquote><blockquote>Are employee X’s current courses covered under their benefit?</blockquote><p>We group these and related questions under a domain that we, at Guild, refer to as <em>eligibility</em>.</p><h3>Eligibility</h3><p>In practice, the aforementioned questions end up looking more like:</p><blockquote>Based on Employee X’s job title, tenure at the company, and cumulative GPA at University Y on November 16th, 2019 — how much money does Company Z owe to University Y to cover Employee X’s tuition for the term starting that day?</blockquote><p>To answer these questions, Guild aggregates data daily from both sides of the marketplace (employer partners and academic partners), and evaluates that data against rules defined as part of that employer’s benefit policy.</p><p>This process of aggregating and persisting eligibility data has gone through a number of iterations since Guild’s inception. Today, our engineering organization consists of over 100 engineers and many squads focused on solving different technical problems, most of which either provide data that is an input to eligibility, or utilizes eligibility data in some way. On the eligibility squad, we set out to create a single source of truth where operations, engineering, business intelligence, and our students can find the answers to these questions.</p><h3>Problem</h3><p>In a single relational database, doing a query that joins data from a few tables with some date filters and boolean logic would be fairly straightforward. At Guild, the data we need to derive eligibility decisions is spread across several different systems and databases. Our goal was to find a way to utilize data from these various platforms to provide current day eligibility information to frontends and student teams, and historical eligibility snapshots to batch processes such as invoicing.</p><h3>Architecture</h3><p>Guild engineers have done some great work over the past year that has enabled us to scale rapidly and decouple systems and teams. One major advancement in our engineering infrastructure was the addition of our <a href="https://aws.amazon.com/kinesis/">AWS Kinesis</a>-powered <a href="https://medium.com/extra-credit-by-guild/get-on-the-bus-adec3a826d58">Event Bus</a>. Several systems from which we needed data were publishing events to the event bus already, so we decided to fully commit to an event-driven architecture and rely on those events to determine eligibility.</p><p>The high-level idea was to listen to events from a bunch of systems, recalculate someone’s eligibility whenever a relevant event occurred that may impact their eligibility, and persist that calculation to an append-only history of eligibility in Postgres. We built this system entirely on AWS using <a href="https://aws.amazon.com/kinesis/">Kinesis</a>, <a href="https://aws.amazon.com/lambda/">Lambda</a>, <a href="https://aws.amazon.com/sqs/">SQS</a>, <a href="https://aws.amazon.com/rds/aurora/serverless/">Aurora Serverless</a>, <a href="https://aws.amazon.com/appsync/">AppSync</a>, with <a href="https://docs.aws.amazon.com/cdk/latest/guide/home.html">CDK</a> as our tool of choice for configuring and deploying infrastructure.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NL-TARrJZOVFZ_YK" /><figcaption>Eligibility Data Persistence Pipeline</figcaption></figure><h3>Components</h3><h4>sink</h4><p>sink is the main entry point into the eligibility processing pipeline. sink listens to multiple Kinesis streams as an event consumer within the event bus. It provides these functions:</p><ul><li>Fan-in events from multiple upstream event publishers</li><li>Filter out events we don’t care about from those publishers</li><li>Take events that we do care about, and normalize data from those events into a standard message format that gets pushed onto our queue for processing</li></ul><p>In sink we defined an abstract class that could be used to implement listeners for various streams. Using CDK for our infrastructure-as-code allowed us to do some cool things like dynamically introspect all Stream subclasses and set up the required AWS resources to enable sink to pull events off those Kinesis streams.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/3f38df4eea5d0dcf2d31b7a6f6ae738a/href">https://medium.com/media/3f38df4eea5d0dcf2d31b7a6f6ae738a/href</a></iframe><h4>queue</h4><p>queue is a streaming collection of eligibility calculation jobs that organizes our workload using <a href="https://aws.amazon.com/sqs/">AWS SQS</a>. When weighing architecture options, we considered driving our workload directly off of Kinesis streams but decided to add this intermediate buffer for the following reasons:</p><ul><li><strong>Observability — </strong>a single queue of normalized messages representing our workload simplified metric collection, dashboard creation, and alerting. We know that invocations of the downstream writer Lambda map 1:1 to jobs on the queue. This makes it easy to trace eligibility calculations as they flow through the system.</li><li><strong>Replay — </strong>having one single queue for our workload with a single associated dead-letter queue made it very straightforward to retry failed jobs by just pushing them back onto the queue.</li><li><strong>Decoupling — </strong>as long as the sink received an event from an upstream and the associated job made its way onto the queue, processing and scaling from that point forward was entirely within our domain and we could tweak our internels as needed without having to change configuration on the centralized event bus.</li></ul><h4>writer</h4><p>writer&#39;s responsibility is to process jobs from queue. writer Lambdas round-robin jobs off queue as they come in, run data from those jobs through the eligibility rules Lambda, and append new eligibility data into the database. If there was a change in a user’s eligibility from the latest one we have recorded, it publishes an event back onto the event bus to notify other systems that are interested in this.</p><h4>poller</h4><p>poller is a Lambda that sits outside of the event bus data stream, and whose job is to invoke an eligibility calculation in circumstances where we don’t receive an event that will trigger one. An example is tenure-based eligibility. Typically we just receive hire date information from our employer partners, and need to make sure that after N days have passed from their hire date, their eligibility is re-evaluated.</p><h4>eligibility</h4><p>eligibility is the brains of the pipeline. It houses employer eligibility policy rules, and evaluates them against user state to provide an eligibility determination with metadata around why output is the way it is.</p><h3>Learnings</h3><h4>Data is hard</h4><p>Our work was only half done once we built the system, deployed it, and backfilled millions of eligibility decisions from 2 years worth of data. The process of testing and QAing data, verifying data integrity both against policy logic and legacy systems that were currently in use, and triaging discrepancies/deploying fixes as we discovered them was long and tedious.</p><p>We developed a couple of data quality dashboards that evolved with the help of our business intelligence team. We also performed a healthy amount of querying in our data warehouse to track down tricky issues from race conditions, upstream bugs, and other things.</p><h4>Time travel is hard</h4><p>If we only had to worry about starting from today and updating eligibility as we marched forward into the future, things would be pretty straightforward. However, it is occasionally the case that we receive updated/corrected historical data and need to make some change to a past eligibility decision we made. Also, other service teams have some internal logic around validating and “promoting” data after it’s received, that we had to account for — it’s not always just “a new event happened”.</p><p>When historical data changes, we still need to keep the previous version of that day’s eligibility around, as there’s a chance someone may have seen it or it may have gotten used in a business process. Doing all this in the context of a database table that is effectively “append-only” proved to be tricky in certain edge cases.</p><h4>Uptime is hard…when you’re a needy service</h4><p>Prior to the full persistence pipeline being complete and data being available in our database, the eligibility API calculated an employee’s eligibility inline during a request. This meant our uptime was the product of all our downstream services’ uptimes, and we were directly impacted by any outages or degradations they experienced. While in this world, we struggled a lot to figure out what we wanted to get alerted about (and how), what we were responsible for versus what issues we delegated to our downstream services to troubleshoot, and how demanding we were on our downstream services in terms of performance and SLAs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NvIqIVg_YK_y5wxN" /><figcaption>AWS X-Ray visualization of Eligibility Dependency Graph</figcaption></figure><p>To ease the addition of new API clients into our codebase while also protecting us from breaking changes, we introduced a Downstream abstraction that required API clients in our code to implement some functionality that automatically added them to our health checks, and reported latency of their requests to a dashboard.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/f1091eff04e70edcb7ebbcb8dea8729e/href">https://medium.com/media/f1091eff04e70edcb7ebbcb8dea8729e/href</a></iframe><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3uvTC76uU6l1pQoP" /></figure><h3>Conclusion</h3><p>While this system is still nascent, we have tens of millions of eligibility records persisted for several million employees and process thousands of events per day to keep our data up-to-date. We have used this data to enable new UIs and scale our backend financial processes.</p><p>In addition to achieving our goal of creating a centralized source of truth for data within the eligibility domain, this processing pipeline and paradigm has since been extended and generalized to serve other use-cases in the organization where persisted, auditable derived/calculated data as a result of events in our ecosystem is a requirement.</p><p>I look forward to seeing this system evolve to meet new challenges and help the business scale as we add more partners on both sides of our learning marketplace.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ebea36256f7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/eligibility-at-guild-adventures-in-event-processing-on-aws-8ebea36256f7">Eligibility at Guild — Adventures in Event Processing on AWS</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Moving our Guild SPAs from Heroku to AWS]]></title>
            <link>https://medium.com/extra-credit-by-guild/moving-our-guild-spas-from-heroku-to-aws-a0ae5973975b?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/a0ae5973975b</guid>
            <category><![CDATA[single-page-applications]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[frontend-architecture]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[software-architecture]]></category>
            <dc:creator><![CDATA[Justin Menestrina]]></dc:creator>
            <pubDate>Thu, 01 Oct 2020 19:06:36 GMT</pubDate>
            <atom:updated>2020-10-01T19:06:36.380Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Migrating Single-Page Applications from Heroku to AWS</strong></h3><p>In the second quarter of 2020, Guild Education began transitioning our Single-Page Application (SPA) hosting stack from Heroku over to AWS. For us, SPAs were only the first major step in migrating all of our deployment infrastructure to the new platform. We chose AWS, despite several robust and fully-featured alternatives like Heroku and Netlify, for the following reasons:</p><ul><li><strong>Cost. </strong>When Guild started in 2016 and had only a handful of applications, cost was not a major factor. But we now have 14+ deployed SPAs, with more coming online each year.</li><li><strong>Flexibility and capacity to scale. </strong>As we scale our teams and capabilities, custom tooling built to align with our needs and use cases enables our business and engineering teams to move faster</li><li><strong>Access control, auditability, and compliance (e.g. </strong><a href="https://en.wikipedia.org/wiki/System_and_Organization_Controls"><strong>SOC</strong></a><strong>). </strong>Serving a broad spectrum of employers, university partners, and students means more responsibility to keep our data safe and secure.</li></ul><h3>AWS Migration Goals</h3><p>After settling on AWS as our new platform of choice, we set several main goals for our re-architected SPA deployment:</p><ol><li><strong>Decouple application <em>hosting</em> from application <em>routing. </em></strong>Today, Guild relies on <a href="https://www.cloudflare.com/lp/overview-l/?&amp;_bt=444210169424&amp;_bk=cloudflare&amp;_bm=e&amp;_bn=g&amp;_bg=104534019073&amp;_placement=&amp;_target=&amp;_loc=1013859&amp;_dv=c&amp;awsearchcpc=1&amp;gclid=CjwKCAjw_NX7BRA1EiwA2dpg0qoFTT4I2Wx64t_HsDl16GfdZMAKFiPuejb4lsGvz-IdMqwK1nRs_RoCllcQAvD_BwE&amp;gclsrc=aw.ds">Cloudflare</a> to handle our routing (e.g. ensuring that when a user hits a url, they land where we want them to). But we saw a future where we could replace Cloudflare with AWS technologies like<a href="https://aws.amazon.com/route53/"> Amazon Route 53</a> and<a href="https://aws.amazon.com/lambda/edge/#:~:text=Lambda%40Edge%20is%20a%20feature,improves%20performance%20and%20reduces%20latency.&amp;text=Lambda%40Edge%20runs%20your%20code,content%20delivery%20network%20%28CDN%29."> Lambda@edge</a>. To allow for this evolvability, we wanted our routing implementation to be replaceable without affecting our hosting infrastructure. To accomplish this, we created an API interface between the routing and hosting layers.</li><li><strong>Establish a standard pattern for how SPAs are deployed and configured. </strong>With our previous Heroku solution, our SPAs had many divergent deployment and configuration patterns. These differences made it harder for engineers to work across teams because of the overhead of needing to learn new patterns. Examples of some differences include: a) Different tooling chains for CI/CD (GitHub actions vs CircleCi); b) Some applications used Cloudflare<a href="https://www.cloudflare.com/dns/"> DNS</a> for routing while others used Cloudflare<a href="https://blog.cloudflare.com/introducing-cloudflare-workers/"> workers</a>; c) Teams had their own URL patterns and webpack configurations for how they handled test/pull request applications. We deemed the migration an excellent opportunity to rethink our frontend deployment patterns and standards and improve developer experience.</li><li><strong>Continue to support Guild’s multiple URL patterns. </strong>Guild serves web applications via two main URL patterns: subdomain-based URLs and path-based URLs. Our new infrastructure had to support both in order to be successful. The subdomain serves as the unique identifier in determining what application a subdomain-based URL serves. Subdomain based URLs take the form<a href="https://subdomain.guildeducation.com/"> https://subdomain.guildeducation.com</a>. Path-based URLs take on the form<a href="https://subdomain.guildeducation.com/path"> https://subdomain.guildeducation.com/path</a>. The path serves as the unique identifier in determining which application gets served. This means different subdomains could serve the same application if the path is the same. For example, we host our programs catalog at <a href="https://subdomain.guildeducation.com/catalog">https://walmart.guildeducation.com/catalog</a> and<a href="https://subdomain.guildeducation.com/catalog"> https://disney.guildeducation.com/catalog</a><strong>, </strong>but both URLs serve the same application code.</li></ol><h3>New Architectural Overview</h3><p>With our platform and technologies chosen, requirements set, and our goals defined, we set out to design a solution that would meet these goals.</p><figure><img alt="Overview diagram of our SPA deployment architecture." src="https://cdn-images-1.medium.com/max/1024/1*uyMNRKLi0LudTeyflxDfIA.png" /><figcaption>Overview of our SPA deployment architecture.</figcaption></figure><p>Our deployment architecture can be divided into two main categories:</p><ol><li><strong>SPA hosting</strong></li><li><strong>URL Routing</strong></li></ol><p>As mentioned earlier, this split comes from the desire to separate our hosting and routing layers. By splitting them up, we are better able to design for future changes.</p><h3>SPA Hosting</h3><p>“Hosting” simply refers to where the SPA files (html, css, and javascript files, etc.) physically live, and a URL that can be hit to access them.</p><p>Our new design uses Amazon Simple Storage Service (S3) for hosting. Each SPA is given its own S3 bucket. Some of the key advantages of S3 are its high availability (99.9%) and robust backups. For any public-facing website, where being available at all times is critical, these features provide major advantages. S3’s biggest drawback is that it can only host static artifacts. Because our SPAs are <a href="https://dzone.com/articles/client-side-vs-server-side-rendering-what-to-choos">client-side rendered</a>, this limitation is ok.</p><p>For URL access, SSL security, and caching, Guild uses <a href="https://docs.aws.amazon.com/AmazonCloudFront/latest/DeveloperGuide/Introduction.html">Cloudfront</a> (not to be confused with Cloud<em>flare</em>, already mentioned, for DNS and routing). AWS Cloudfront provides us with a unique URL that can be used to access the application hosted in a bucket.</p><p><strong>Tips for S3 hosting</strong></p><ul><li>Keep separate buckets for each SPA. Doing so will help keep each SPA isolated and allows for independent deployment lifecycles.</li><li>Enforce naming standards to enable discoverability. You want it to be fairly obvious what a bucket does, just by looking at the name. We name our SPA buckets after the Github repo they serve, and the domain name they are hosted on. (E.g. {github-repo-name}.{domain-name}.com). This allows us to easily identify which application a bucket hosts.</li><li>Remember to make use of <a href="https://docs.aws.amazon.com/AmazonS3/latest/dev/UsingMetadata.html">object “metadata”</a>. We use metadata to provide HTTP headers like <a href="https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/MIME_types">MIME types</a> and <a href="https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Cache-Control">cache control headers</a>.</li></ul><h3>URL Routing</h3><p>URL routing emerged as the most challenging aspect of our deployment architecture. It also proved to be the most vital in accomplishing the decoupling we were going for.</p><p>Previously, each application employed its own unique way of routing its public-facing URL (i.e. the route a user visited to see a page) to the URL that Heroku provided for an application deployment.</p><p>To address these challenges, we broke up URL routing into two parts:</p><ol><li><strong>Translating a public request</strong></li><li><strong>Translating an asset request</strong></li></ol><h3>Translating a public request</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vIavlZYLSg9wtCyDbVm-pg.png" /><figcaption>Diagram showing that a user-friendly url is translated into a more complex Cloudfront url that points to an S3 bucket</figcaption></figure><p>A public request refers to the process of translating a public URL into the Cloudfront URL that the requested application is served from.</p><p>To better understand how this works, let’s walk through the following scenario:</p><p>I am a user who visits <a href="https://walmart.guldeducation.com/programs-catalog">https://walmart.guildeducation.com/catalog</a>. The code that this URL serves lives in a Github repository called programs-catalog. How does that source code make it to my browser when I visit <a href="https://walmart.guldeducation.com/programs-catalog">https://walmart.guldeducation.com/catalog</a>?</p><ol><li>A Cloudflare worker translates the public URL <a href="https://walmart.guldeducation.com/programs-catalog">https://walmart.guldeducation.com/catalog</a> into a<strong> standardized internal URL:</strong> <a href="https://programs-catalog.guildeducation.com/">https://programs-catalog.guildeducation.com</a></li><li>Next, a Cloudflare DNS entry translates the standard internal URL <a href="https://programs-catalog.guildeducation.com/">https://programs-catalog.guildeducation.com</a> to the correct Cloudfront URL in AWS</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FLN9cyuYYfBDfDRrxklokQ.png" /></figure><p><strong>The core idea is that we created a standardized internal URL that uniquely identifies a Github repository.</strong> We settled on the format https://{github-repo}.guildecucation.com. For this example, the standardized internal URL is <a href="https://progams-catalog.guildeducation.com/">https://progams-catalog.guildeducation.com</a></p><p>We chose this pattern because Github repos are unique within orgs, and it’s an easy way to identify which repository a URL refers to.</p><h3>Translating an asset request</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A2fXYUF-udOs7W2ysM3GYw.png" /></figure><p>Continuing with our example from above, once <a href="https://walmart.guildeducation.com/program-catalog">https://walmart.guildeducation.com/catalog</a> has been translated and its index.html page has been returned to the user, the user’s browser now needs to get the static assets that are required to render the page.</p><p>A static asset is a file that is loaded via a link or script from an HTML document. Examples of assets include javascript files, css files, and images.</p><p>Because we need to support path-based applications like in this example, we need to use absolute links for our static assets.</p><p><strong>The key insight is that we can reuse our standardized internal URL to create an absolute link for all our assets</strong>. As demonstrated above, the Cloudflare DNS record correctly translates that internal URL to the Cloudfront URL. By utilizing this strategy, we are able to standardize the way all of our SPAs create links to their assets.</p><h3>Putting it all together</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*z89TTe4PBf8SqnY3pWi-Wg.png" /></figure><p>The solution outlined above accomplishes several of the architectural goals we set out to achieve.</p><h4>Decouples routing, application builds, and hosting infrastructure</h4><p>We created two primary interfaces that serve to decouple the hosting and the routing layers:</p><p><strong>Standardized Internal URL</strong></p><p>This serves as a way for the routing to map assets and public URLs back to the correct DNS record for an application. Applications need to know only about this URL. Because it’s simply a standard, it can be reused even if the routing layer is replaced.</p><p><strong>Cloudflare DNS record</strong></p><p>The hosting infrastructure automatically creates a DNS record in Cloudflare when it’s created. The consequences of this interface means that if the routing layer gets replaced later, we only need to change where the hosting infrastructure creates this DNS record.</p><h4>Establish a standard pattern for how all SPAs are deployed and configured</h4><p>All applications, regardless of their public URL requirements, are configured in the same standardized way: a standardized internal URL representation. This creates better uniformity across teams, which helps increase velocity as we scale.</p><h3>Conclusion</h3><p>Our migration from Heroku to AWS enabled us to redefine best practices and standards in how we deploy SPAs. Our utilization of AWS and Cloudflare fit our needs well, but isn’t necessarily recommended for everyone. Most existing patterns for hosting SPAs in AWS recommended using all AWS services. If you’re able to do this greenfield, I’d recommend using AWS-first technologies for DNS and routing, like S3 and Edge.</p><p>However, hopefully this pattern serves as a guide for other teams or companies that are looking for migration strategies to AWS and ways to create a more decoupled SPA deployment architecture.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a0ae5973975b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/moving-our-guild-spas-from-heroku-to-aws-a0ae5973975b">Moving our Guild SPAs from Heroku to AWS</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Scale for Evaluating Mobile Responsiveness]]></title>
            <link>https://medium.com/extra-credit-by-guild/a-scale-for-evaluating-mobile-responsiveness-b00226743924?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/b00226743924</guid>
            <category><![CDATA[responsive-design]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[mobile]]></category>
            <category><![CDATA[mobile-app-development]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[Alex Tzinov]]></dc:creator>
            <pubDate>Thu, 24 Sep 2020 14:30:59 GMT</pubDate>
            <atom:updated>2020-09-24T14:30:58.923Z</atom:updated>
            <content:encoded><![CDATA[<h4>Exploring a new way to measure a product’s progress toward “responsiveness”</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sjURTBQp1O4SvJYq" /></figure><p>The terms “mobile friendly” and “responsive” are far from being a binary measure of whether “you have it” or “you don’t.” Defining mobile friendly on a more granular scale is crucial if we want to improve our mobile experiences. <em>If you can’t measure it, you can’t improve it. </em>This blog post explores a new way to measure mobile web maturity and assesses where Guild Education’s product stands in terms of our own progress.</p><p>At Guild, we strive to serve an ever-growing diversity of students. We recognize that our users have varying levels of computer/phone savviness as well a wide range of potential devices (laptops, work computers, slower smartphones, larger smartphones, etc). One of the ways we navigate this varied landscape is by ensuring that our student-facing web apps are <strong>responsive. </strong>That they’re <strong>mobile friendly</strong>. That we design with a <strong>mobile-first </strong>approach. These buzzwords encompass a lot of things:</p><ul><li>A web app that looks good regardless of device size</li><li>Starting designs small and then growing to wider screens</li><li><a href="https://bradfrost.com/blog/post/7-habits-of-highly-effective-media-queries/">Choosing breakpoint values to serve the content</a>, <a href="https://gist.github.com/basham/3b24062dfaecaa712a68">not device dimensions</a></li><li>Faster load times</li><li>Call-to-action buttons that are easy to find and click</li></ul><p>These guides and rules are important, and it’s common knowledge that tolerance is dwindling amongst mobile users for slow and clunky sites. Everyone knows that mobile is essential, but not everyone knows what that means exactly. I felt that the concept lacked a skeletal foundation on which to place things such as font sizes, breakpoints, accessibility, mobile performance, and more.</p><p>I settled on representing mobile design (and its corresponding technical implementation) as a <strong>spectrum of solutions</strong>, ranging from completely ignoring mobile to making it the number one focus. Each phase has distinct characteristics that separate it from the previous one.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Hn-SJJr39YaiPKmx" /><figcaption>Modeling the mobile design landscape as a growing progression from no mobile solution to a performant and very usable solution</figcaption></figure><p>Every piece of literature around mobile design — and likewise, every app that strives to implement all the buzzwords of mobile design — falls somewhere on this spectrum.</p><p>We’ll dive into the phases of this spectrum, assess where Guild lies on it, and address actionable next steps to move our product towards mobile enlightenment.</p><p>The spectrum itself:</p><ul><li><strong>Ignore: </strong>You built your app pretending mobile users don’t exist. What it looks like on mobile is anyone’s guess. 1920 x 1080 or bust.</li><li><strong>Acknowledge: </strong>You anticipated someone might use your app on mobile. You add a “Coming Soon” page or redirect when you detect mobile. Done.</li><li><strong>Handle: </strong>You tested your app on a few mobile devices. It’s not a complete disaster. Good enough for now, “it’s usable.”</li><li><strong>Consider: </strong>You made intentional design decisions to factor in mobile: layout breakpoints, adjusted font sizes, alternative navigation methods.</li><li><strong>Optimize: </strong>You made mobile a priority: you made call-to-actions easier to find, leveraged native inputs, and minimized user journey interactions.</li><li><strong>Supercharge: </strong>You made mobile <em>the </em>priority: you reduced bundle sizes, scrutinized load times, tested extensively, and provided native solutions.</li></ul><p>Where a product lands on this spectrum is determined as much by <em>technical</em> factors (skills of developers, technology choice, etc) as it is by <em>product and business factors </em>(available bandwidth, company priority, user base, etc). There isn’t necessarily a bad solution — although some awareness of your mobile status is always preferred to none — and the magical end goal will likely be different from company to company. What’s important is recognizing where you stand and being able to identify the actionable next steps to move forward.</p><h3>Ignore</h3><p>While its position at the beginning of the spectrum might suggest that a product that <strong>ignores</strong> its mobile solution is doing something bad, there are some real use cases for a company not to waste mental energy thinking through mobile solutions.</p><p>At my previous company, we built software for security analysts to monitor and react to network events. Our apps were saturated with overlapping configuration panels, complex charts and graphs, notification alerts, and tables with thousands of rows. Thinking that our solution could exist on mobile would be foolish. It wasn’t part of our product offering, and no customer would ever seek it. Ignoring mobile was acceptable.</p><p>The luxury of ignoring mobile exists for products that have a very specific user segment and don’t risk losing customers by not offering a mobile solution. So you either know that your product isn’t intended for mobile (i.e. acute customer segment awareness) or you have some idea of where you stand on smaller devices.</p><p>Mint is an example of a product that has a fully-fledged native app for iOS and Android, but has decided to <strong>ignore</strong> their mobile web app completely.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*RiaOTtDb0Q_oBa8t" /><figcaption>Mints desktop experience dropped into a mobile device</figcaption></figure><p>Mint allows a user to log in on mobile and just renders their full desktop app. They <strong>ignore </strong>their mobile solution; it’s less than ideal and not very usable. They invested their resources into a native app and chose not to do anything about users trying to log in to their web app on their phones. Catastrophic? Certainly not, but we’ll see a different way of handling this same use case in the next phase.</p><h3>Acknowledge</h3><p>Acknowledging that having mobile users on your app is a <em>possibility </em>is enough to motivate at least some kind of minimal “solution” that could take various forms.</p><p>One is to gracefully handle mobile sizes with a landing page that simply informs users that you don’t support mobile:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/0*j_cpEkIEeghuoh4g" /><figcaption>It’s not pretty, and most certainly a copout, but with just a few @media queries and some fancy “Please stick with us while we develop mobile” type branding, this solution could be a quick and dirty way to retain users</figcaption></figure><p>Another option, often called adaptive design, is to redirect the user to a separate, mobile version of your site if it is detected that they are on a mobile device (m.mysite.com vs mysite.com). Mozilla’s Developer docs recommended against this, calling out the error-prone nature of detecting user devices server-side, as well as the double maintenance of having two different code bases.</p><p>Yet another option — if the product has a native app and doesn’t want to build a mobile-friendly browser version — is to have a popup telling the user about the app.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NGgp3cJvLMKyImDa" /><figcaption>Robinhood’s very intentional mobile web experience acts as just a landing page and a pointer to the native app</figcaption></figure><p>Unlike what we saw earlier with Mint (which also has a native app), Robinhood actually <em>prevents </em>a user from signing in at all on mobile and only renders an informational landing page with a popup about their app. They <strong>acknowledge </strong>that users might hit their site before their app, and use their site to inform and then direct users towards their app.</p><h3>Handle</h3><p>At this point, your team has made an intentional decision to put enough effort into the mobile solution to make it more tailored and usable. It’s not pretty, resources aren’t heavily allocated, and testing is barebones, but the app is usable and adjusted for mobile. Perhaps you ship a heavily slimmed-down feature set, or you simplify your use cases to “desktop” and “mobile” and don’t provide the granularity of tablet and older smartphones. You haven’t invested enough resources to establish reusable patterns or redesign every feature, but you have ensured the critical portions of the app fit and that the experience isn’t one of obvious neglect.</p><h3>Consider</h3><p>At this phase, you have placed a much heavier emphasis on mobile use cases. Key things are implemented like increasing font sizes, removing horizontal scroll, and restructuring navigation. Every feature also has some minimal specs for mobile implementation. Horizontal dropdown navigation isn’t just resized to fit across a mobile screen; it’s replaced with side navigation that stays tucked away. Inputs are made bigger to attract fingers more easily, and paddings are adjusted to make content more breathable.</p><p>Developers are likely to be using the mobile inspect mode in browsers allowing them to test their UIs on several screen sizes quickly. Developers are also laying groundwork for common patterns and reusable UI infrastructure to establish mobile consistency. Breakpoints are consistent across the app, fonts readjust in unison, and behaviors like buttons becoming full-width and sticking to either the top or bottom of the screen are commonplace. <strong>Considering</strong> the mobile solution is just as much about design decisions as it is technical investment in reusability and consistency.</p><h3>Optimize</h3><p>At this phase, you’re no longer just making sure the content fits, but that you have <em>the right content.</em></p><p>You can always make a table fit and be navigable by scrolling, but is a table the <em>best way to present data</em> on mobile? Yes, a user can scroll to the top to always find the submit button, but will the user <em>want to scroll every time? </em>You can easily save on space by making extra content collapsible, but is that extra content <em>even necessary?</em></p><p>No longer just the layout, the <em>user journey</em> is now also redesigned through the lens of someone with:</p><ul><li>two thumbs (and many of us even prefer using our phone with <em>one</em> hand)</li><li>drastically shorter attention span (we’re often distracted when on our phones)</li><li>potentially less time to achieve a task (we rarely spend several minutes trying to do one thing on our phones, although exceptions to this exist)</li><li>a trained expectation of performance (other fast mobile apps like Google Maps and Instagram set this bar high)</li></ul><p>With this new persona in mind, a lot has to change. Redesigning existing content to be presented differently, removing/consolidating steps in the user flow, streamlining any longer tasks with animations and transitions to keep ever-fleeting user attention, and putting call-to-action buttons front and center.</p><p>Things like information <em>density </em>are taken into consideration. A user on their phone <em>may</em> not expect encyclopedia-level information; they may want the bare minimum needed to accomplish their task and move on. Other users who don’t have access to a computer, and a phone is their only way to view information, might actually want that detailed wall of information.</p><p>Optimizing for mobile means reconsidering the breadth of users you have (first time users, power users, desktop-less users, etc.) and designing for each <em>within the realm of mobile</em>. The user personas that are kept in mind during architecture, design, and product meetings now must also include permutations for mobile devices. Every pixel (copy, buttons, animations) is scrutinized and intentionally placed given the limited screen real estate, and entire experiences are designed specifically for the mobile user.</p><p>Diagrams such as the one below are kept in mind to optimize for finger placement connivence. This may seem overkill, but if you are optimizing for the mobile user, every detail adds up.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/957/0*2xuHqMBA_vj955Kq" /><figcaption><a href="https://www.smashingmagazine.com/2016/09/the-thumb-zone-designing-for-mobile-users/">https://www.smashingmagazine.com/2016/09/the-thumb-zone-designing-for-mobile-users/</a></figcaption></figure><h3>Supercharge</h3><p>At this point, mobile is <em>the </em>focus, and it goes way beyond just building responsive design layouts that fit nicely. Performance, load times, and network latency are all now crucial. A site cannot achieve mobile superiority if that super slick, fluid, well-animated app takes 9 seconds to load on a user’s device.</p><p>At this phase, a substantial amount of time and effort is spent on <em>speed. </em>Things like bundle size, the number of requests a UI has to make to the server (each one incurring network latency penalties which can add up), and overall CSS paint time and JS processing time become important metrics. Testing occurs on a plethora of devices.<a href="https://developers.google.com/web/tools/lighthouse"> Lighthouse</a> (a Chrome performance audit tool) is used to simulate slower processing power and 3G speeds. Caching strategies and bundle optimizations like code splitting and tree shaking are must-haves. Content that gets downsized <em>visually </em>(images, banners, logos, icons) to fit on mobile should also be downsized <em>physically </em>(in terms of bytes) to help reduce load times.</p><p>It wouldn’t be a stretch to say that keys to achieving this performance level include CI checks that ensure bundle sizes and page load times remain below certain thresholds. A code change that drastically increases loading time is blocked for further review. As mentioned earlier, mobile is <em>the </em>focus, and everything from the release process to server architecture to content delivery networks is optimized with mobile speed in mind.</p><p>Responsive design isn’t complete just by resizing/reshaping your page content and optimizing for thumb access. Responsive design should be seen as <em>encompassing </em>design. How wide of an audience can your product serve? With performance bottlenecks, this audience shrinks. Page content doesn’t matter if the user never sticks around to see it.</p><p>So now comes the big question: where are we at Guild?</p><h3>Where Are We at Guild</h3><p>We will dive into Guild’s various mobile aspects as they stand today, but right out of the gate, one of the biggest hurdles we currently have is <strong>performance.</strong></p><p>Performance is not necessarily measured in bundle size, download times, number of caches, Time To First Paint’s (TTFP), or animation speeds. While these are all important metrics, they don’t necessarily represent the user journey. Our user journey can (and in my opinion should) be seen as the time spent from the moment a user enters the funnel to the moment they have achieved or (nearly achieved) some measurable metric, such as <strong>application start rate.</strong></p><p>Currently, one particular user journey at Guild can include up to four steps:</p><ol><li>User loads and then signs in from our <em>Partner Page</em> (information about Guild’s educational benefits)</li><li>User is then redirected to our <em>Student Home Page </em>(launching spot for logged-in users)</li><li>User then navigates to the <em>Catalog</em> (our list of academic programs) to search for programs</li><li>User finds a program they’re interested in and clicks <em>Apply</em> (Application page)</li></ol><p>Users will spend a minimum of <strong>24</strong> <strong>seconds*</strong> waiting on those 4 pages to load. Not 24 seconds exploring, thinking, navigating, scrolling through options, becoming informed, learning about Guild. 24 seconds of <em>loading time</em>. That’s over a WiFi connection on a MacBook Pro.</p><p>A <strong>supercharged </strong>mobile experience is impossible if a user spends at least 24 seconds staring at loading pages in a world where <a href="https://time.com/12933/what-you-think-you-know-about-the-web-is-wrong/">average web page visit times are 15 seconds</a>.</p><p>Which brings me to Guild’s position on this spectrum: In my opinion, we are currently in the process of <strong><em>optimizing.</em></strong></p><p>We are well past simply considering the mobile use case. We have done research to collect the most often used devices. We test extensively on multiple devices (and even have a device cart at the office containing older smartphones and tablets). Newer features have designs uniquely tailored for mobile experiences, and navigation methods exist for both use cases (desktop and mobile).</p><p>Are we perfectly optimized? This fluctuates and differs across the organization, but there is always room to improve. And are we supercharged? Not yet. The 24-second, 4-page journey is simply too convoluted and slow to be considered a supercharged experience, and we haven’t really tackled our bundle sizes in an intentional manner.</p><h3>Next Steps</h3><p>There are a lot of ways our organization hopes to move towards the supercharged state. We’ve already identified some lower hanging fruit, such as increasing our Cypress testing and implementing certain webpack strategies, and we know that the growing number of separate React apps have to be eventually be consolidated.</p><p>Stay tuned for a much deeper discussion into that exact question — how do we get from <strong>optimizing </strong>to <strong>supercharged </strong>— where we will dive into technical details, architectural decisions, and ways to navigate the business/product aspects. Part 2 of this blog post is coming soon.</p><p>Until then, we’ve seen how mobile/responsive design is better represented by a spectrum than by a black-or-white evaluation. Every organization is somewhere on this spectrum, and improving your offering and ultimately serving more users requires a strong understanding of the current state of your mobile product.</p><p>*5 rounds of tests were done recording events on the network tab and measuring perceived time (subtracting time spent moving mouse to navigate) during the navigation of those 4 pages</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b00226743924" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/a-scale-for-evaluating-mobile-responsiveness-b00226743924">A Scale for Evaluating Mobile Responsiveness</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Life Lessons in Engineering Leadership]]></title>
            <link>https://medium.com/extra-credit-by-guild/life-lessons-in-engineering-leadership-e6c74b91a1a5?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/e6c74b91a1a5</guid>
            <category><![CDATA[guild-education]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[culture]]></category>
            <category><![CDATA[trust]]></category>
            <category><![CDATA[engineering-management]]></category>
            <dc:creator><![CDATA[Jay Ockers]]></dc:creator>
            <pubDate>Thu, 10 Sep 2020 13:12:57 GMT</pubDate>
            <atom:updated>2020-09-10T13:12:57.818Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VHpWq6IvqvheucnM" /><figcaption>Photo by <a href="https://unsplash.com/@austindistel?utm_source=medium&amp;utm_medium=referral">Austin Distel</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Trust is the foundation of good leadership. In my fifteen years as a software engineer and seven years as an engineering manager, I have seen leaders who failed to inspire trust, and I have failed in many of the same ways. Only recently have I really focused my practice toward emphasizing trust as the core of my relationship with my reports, and I’ve built stronger relationships with my team as a result. I hope that you too can learn from my observations and find some inspiration for your own leadership practices.</p><h4>Transitioning from Engineer to Manager</h4><p>Early on in my career as a software engineer, I struggled to understand the role managers should play. I expected them to be more involved in technology, and I often felt disappointed by their lack of current knowledge and mentorship. Later, I encountered managers who were too involved in technical decisions, causing them to fail as leaders. In my own practice, I too have found myself leaning too far towards one extreme or the other without explaining my intentions to my teams.</p><p>In my first role managing engineers, I spent too much time in code and troubleshooting and not enough time focused on my team. This felt natural because it was how I had always wanted my managers to behave. After a year on the job, however, I realized that my choices had left my team without the leadership that they needed to advance their careers. Recognizing this, I started to rebalance my workload to focus on creating meaningful growth opportunities for my team.</p><p>I gradually stopped writing code and reviewing every pull request and started focusing on one-on-ones with my engineers, strengthening my feedback loops and creating a dedicated space for career growth. I took over handling site reliability requests to keep my team focused on product feature work, and I included them earlier in the problem solving process for particularly complex requests. Throughout the year, I evolved my view of my role from focusing on what I contributed to caring about how I supported my engineers.</p><p>A crucial learning from this transition was that I had to be transparent about my intentions with my direct reports. On teams where I was most effective, everyone knew my limitations and understood the role that I intended to play. On teams where I failed to properly communicate how I viewed my role, my engineers grew frustrated as they looked to me to be an expert in our core technologies. Looking back, I believe I would have been more successful if we had a shared understanding of my role and accountabilities.</p><p>In my most recent incarnation, I have felt most effective when I share my limitations and intentions with my teams. Knowing why I choose to be involved only at the abstract level helps them understand how they can best utilize my knowledge. Limiting my time spent on engineering work in this way gives me the flexibility to maintain a balance between my responsibilities as a manager and as an engineering mentor. This is very much a learning process though, and I rely heavily on feedback from my engineers to understand how my efforts impact them.</p><p>Soliciting this kind of constructive criticism from anyone is a challenge, but without trust I can’t imagine it would even be possible. I often struggle to give folks confidence that I hear their concerns and that I am accountable to them. To demonstrate this accountability, I refer to notes that I take during retros, performance reviews, and one-on-one conversations about my commitments. More often than I like to admit, I discover that I am falling behind on these commitments. But acknowledging these lapses and being open about my flaws helps my team understand my limitations.</p><h4>Building Trust</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*gEzS_wu-dCRlirpl" /><figcaption>Photo by <a href="https://unsplash.com/@zmachacek?utm_source=medium&amp;utm_medium=referral">Zdeněk Macháček</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>All of my frustrations with former managers and many of the grievances that my own teams have had with me stemmed from a lack of trust. It is the foundation upon which my most successful teams have been built and something I try very hard to establish and maintain as I lead my teams today.</p><p>I encourage you to look for the humanity in your engineers so that you can understand what motivates them and keeps them engaged with their work. I spent my time as a software engineer frustrated by bosses who thought that managing engineers was impossible because we were too difficult to understand. In my own practice, I have learned to accept that my engineers have hopes, fears, ambitions, and insecurities just like everyone else.</p><p>To identify these characteristics, I rely heavily on the pyramid of needs to understand how to build on the foundations that my teammates already have in place. Sometimes motivation comes from simply encouraging their physical well being through comfortable working conditions, nutritious snack options, and healthy work life balance. If that’s already been addressed, I look to ensure they feel safe in their work environment. In the case of engineers, safety is often more about emotional security. I try to make sure my engineers feel safe sharing their ideas with me and with their peers.</p><p>Encouraging this kind of safety translates well into helping my team members establish strong relationships with their peers. This is an area often overlooked by engineering managers who don’t realize that their more introverted teammates still want opportunities to socialize with their peers but might feel intimidated by large gatherings. Creating opportunities for all personality types and being transparent about your intentions helps folks understand your efforts and give you feedback. In turn, this feedback can help you tailor your efforts to better serve your whole team instead of just the extroverts.</p><p>These events can also be useful for helping your engineers feel appreciated and gain insight into what kinds of encouragement their peers need to thrive on the team. I also try to focus my feedback on recognizing how my engineers’ efforts are impacting the business and how I see their progress unlocking future opportunities. My intention is to build their self esteem and also show them a path forward at all times.</p><p>Once I understand what motivates my engineers, I try to demonstrate my trust by encouraging them to share in the ownership of decisions. I have often felt frustrated by managers who believed that decision making was a function of their role and not something that could be delegated. My frustration came from being asked to build solutions without sufficient context or input behind the decision making process. To overcome this, I try to encourage my engineers to connect with the problems they’re solving by giving them a seat at the table where decisions are made. As a manager, I expect to be held accountable for my decisions, and I trust that when my team is involved in the process, they will stand behind me and support me in a deeper, more meaningful way.</p><p>As my engineers take ownership of their work, I try to stay connected with them by participating in team story grooming sessions and code reviews. My intention here is to give my team insight into my level of understanding and to have more opportunities to celebrate their efforts and offer pointed guidance. As an engineer, I often felt disconnected from my managers when I was creating systems because they didn’t have time for me to explain what I was building. This left me lacking trust that they were good stewards of my career. As a leader, I encourage my engineers to share their struggles and accomplishments with me. This gives me a deeper understanding of the problems they’re facing and how they’ve overcome them. I then use these insights to encourage their growth and better understand where they might be best deployed on future assignments.</p><p>My hope is that forming a mutual understanding with my team will help us collaborate about the nature of their work and what enhances or reduces their productivity. I’ve had many managers who felt it was their role to dictate processes to the team, but I always struggled to trust someone to make these decisions when they didn’t have to live with their impact. I ask my teams to help me define the processes that will work best for them. I start by sharing why I think the status quo is broken so that they know what I hope to accomplish. I also encourage them to discuss their goals and how a new process could impact their work. This builds a sense of shared ownership and empathy toward my needs as well as those of their peers.</p><h4>Sharing Leadership Roles</h4><p>Sharing the burden of leadership with your team can demonstrate your trust in them and ensures the team is getting the best guidance available. I break leadership into three roles: Coach, Captain, and Cheerleader. The Coach is responsible for strategic decisions and long term skill growth. The Captain is responsible for leading the team through tactical decisions and technical skill development. The Cheerleader is responsible for recognizing accomplishments, offering encouragement and reinforcing learning.</p><p>An example of delegating the Coach role is when I ask my lead engineer to mentor junior engineers on a technology. It’s important to me that I make sure the team knows that this takes time away from my lead engineer and also that I feel the less experienced engineers will benefit from more current and deeper technical knowledge that I expect my leads to possess.</p><p>For the Captain role, I look for both technical and non-technical opportunities for folks to take the lead. An example of delegating a technical leadership role is when I ask my senior engineer to design a system and explain their design to the rest of the team. My expectation is that the engineer will lead the implementation effort as well. As the project progresses, I focus on helping them delegate and dedicating time to help folks level up.</p><p>When I delegate the Cheerleader role, I try to find someone who is willing to hold their peers accountable in a constructive and encouraging manner. Within a Scrum Agile team, this can be as simple as delegating the scrum or retro leader role to an engineer. Sharing this role develops empathy amongst the engineers for their peers’ contributions. It also fosters awareness of the effort required to keep projects on track.</p><p>My goal in delegating these roles is to help my team understand how I view leadership. This gives my team a strong foundation to assess my leadership and help me understand where I’m falling short. I don’t expect my engineers to take the reins at every opportunity but I want them to know that I trust them when they do.</p><h4>Key Takeaways</h4><p>I leave you with a few hard-earned lessons. Build up your people by investing in their growth, empowering them to make decisions, personally connecting with their wins and struggles, and giving them ownership of the processes that define their work life. Advocate for intentional leadership and model it for your team so that they can help you carry the burden as Coaches, Captains, and Cheerleaders for their peers. Be mindful of the role you intend to fill on the team, and share that with your engineers. Knowing your intentions helps them shape their expectations for technical mentorship and career guidance. Finally, encourage your engineers to give you feedback. Only by hearing their perspectives can you gauge how well you’ve communicated your intentions and how your efforts are being received.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e6c74b91a1a5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/life-lessons-in-engineering-leadership-e6c74b91a1a5">Life Lessons in Engineering Leadership</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A CMS Case Study: How Our Implementation Allowed Us to Scale for Students]]></title>
            <link>https://medium.com/extra-credit-by-guild/a-cms-case-study-how-our-implementation-allowed-us-to-scale-for-students-8f5899c0d032?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/8f5899c0d032</guid>
            <category><![CDATA[cms]]></category>
            <category><![CDATA[contentful]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[content-management]]></category>
            <category><![CDATA[guild-education]]></category>
            <dc:creator><![CDATA[Jalena Taylor]]></dc:creator>
            <pubDate>Thu, 03 Sep 2020 14:16:20 GMT</pubDate>
            <atom:updated>2020-09-03T14:16:20.271Z</atom:updated>
            <content:encoded><![CDATA[<p>The tech industry is saturated with acronyms that are thrown around colloquially on a day-to-day basis. One acronym that I’ve found to be less commonplace in an engineering org happens to be one of my favorites: <strong>CMS.</strong></p><p>CMS stands for Content Management System, which, at its most basic level, is software that helps you manage your content, whether it’s text, images or more. It’s one of my favorite acronyms because it allowed me and my engineering counterpart to automate ourselves out of the problem we used to work on every single day. We were able to do so because a CMS allows non-developers to make changes to web or mobile applications in an easy, streamlined fashion — no code deploys necessary. We’ll talk more about our CMS case study soon and the immediate return we saw in productivity and scalability.</p><p>When we talk about Content Management Systems, it’s important to note that there are two different types that solve very different business cases:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/903/1*BDK2aT7lOAf5vnL-jCIICw.png" /><figcaption>A visual representation of a Traditional CMS, tightly binding the frontend to the backend, vs. a Headless CMS, delivering content to several frontend applications via API. Image owned by Contentful.</figcaption></figure><ol><li><strong>Traditional CMS: </strong>Your WordPress / Squarespaces of the world. In a traditional CMS, everything comes in one package. Your frontend is tightly bound to your backend. Customization is more difficult, but scaling a website quickly without technical resources is much easier if you’re willing to play in existing templates and sacrifice custom functionalities. A <strong>traditional CMS</strong> is perfect for a project where developer resources may be harder to find, and organizations can benefit from SEO and security plugins that help them get a website off the ground quickly.</li><li><strong>Headless CMS: </strong>An API-based CMS that delivers your content to be consumed as needed by your frontend. A <strong>headless CMS</strong> is perfect for a project that aims to streamline content management in an existing application without sacrificing customization or flexibility. Since it’s completely detached from your frontend experience and delivered via API, plugging an existing application into a headless CMS can be an extremely fast solution to optimize productivity.</li></ol><p><strong>Why We Invested in a CMS: A Case Study</strong></p><p>At Guild Education, we have a mission to unlock opportunity for America’s workforce through education. With 88 million working adults in need of higher education or upskilling in order to compete in the future of work, being able to scale our business to meet the needs of these students is a top priority.</p><p>When I joined the Guild Engineering team a year ago, we were managing a collection of landing pages, each used by one of our employer partners to teach employees about their education benefit.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s8W3TtlCn8yRdJ78S5zA2g.png" /><figcaption>An image of the landing page hosted for Taco Bell employees.</figcaption></figure><p>These pages weren’t doing anything fancy from a technical standpoint (consider them basic HTML marketing pages), but our process for managing them was highly manual. Every time a partner requested an update to these pages, the request had to be fielded by our marketing team, funneled to our product manager who gathered requirements and scoped the task, estimated by engineers, assigned to a sprint, coded, tested, approved, and finally deployed. All for a <em>text change.</em> Working its way through this process could take anywhere from a few days to a week, which wasn’t something we considered an acceptable turnaround.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*FmZopASs9sM3zjQF" /><figcaption>A diagram of the process to make a text change, from partner request to completion. Don’t worry about what the bubbles say, just know that each bubble is a step in this arduous process.</figcaption></figure><p>There was definitely a time in our business this process made sense, but as a company focused on scaling to support as many students going back to school as possible, it was clear that we needed to streamline this flow and free up development resources from making such minimal changes.</p><p>Enter the CMS. We wanted our solution to plug directly into an existing Ruby on Rails application, which had a hand-rolled coupling between our backend and our frontend. Knowing that we could easily break these apart and utilize our existing frontend functionality, we quickly prioritized a headless CMS that would work with our existing infrastructure. Some of our top priorities were:</p><ol><li>Compatibility</li></ol><ul><li>It had to support a variety of frameworks. Our landing pages were in Rails, but our business supports many different types of applications. We wanted to be able to adopt this across multiple teams once we had a proven concept.</li><li>It had to coexist with our analytics and A/B testing implementations</li></ul><p>2. Flexibility, flexibility, flexibility.</p><ul><li>We needed the ability to model our content on demand and be able to adjust easily as business requirements changed</li><li>The CMS had to support a variety of layouts and templates as needed by our partners</li><li>We wanted to release the implementation in a tiered process to create incremental change immediately, instead of having to front-load all of the feature work</li></ul><p>3. Scalability</p><ul><li>We wanted this decision to support our business through a phase of rapid growth. We didn’t want to have to go back to the drawing board in 6–12 months if the CMS we chose could only support us to a certain point</li><li>We had to be able to re-use our content in multiple places. If the same module is present for every partner, we wanted to be able to create that content once, and edit it once when it changed.</li><li>We had to be able to launch a new partner in 1 day or less</li></ul><p>4. Usability</p><ul><li>Our solution needed to be usable by non-developer teammates without an extensive amount of training.</li></ul><p>After thoroughly vetting a large variety of CMS providers, which included spinning up applications in the different interfaces, running through countless demos, and wading through a wide variety of sales pitches, we landed on <a href="https://www.contentful.com/">Contentful</a>. We loved the strong developer documentation, the ease of re-using content across pages and platforms, and the ability to create robust roles and permissions so we could hand off content updates to a non-developer team.</p><p>Our initial implementation took 2 weeks to get up and running. After integrating Contentful in our codebase, we increased our new partner launch velocity from ~2 weeks to ~2 hours. Additionally, small content changes could be made in less than 5 minutes, as opposed to the 1–2 week process bottleneck that those changes experienced previously.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*1lrDvTqrMsAnu7RO" /><figcaption>A diagram of the process to make a text change with our new CMS, from beginning to end. Far fewer bubbles than the version above!</figcaption></figure><p>In the months since that initial launch, we continued to expand upon our feature set. We created all of the fail-safes needed to off-board this product from the engineering team (read: we figured out how to break it a million times and fixed it a million more). We implemented multiple automation features so that all of the necessary pieces could be in place in our ecosystem without tapping into engineering resources at all. A partner could officially launch without a single engineering ticket or a single code deploy.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*lu7hksGwTojwdqog" /><figcaption>A peek into the Contentful UI.</figcaption></figure><p>You may be asking, what do we do with all of our free time now? We’re onto the next project to work toward the goal of unlocking opportunity for America’s workforce through education!</p><p>So what can this mean for you if you’re considering a CMS for your company or team? If you’re sold after what you’ve read, fantastic! I’d love to hear about what a CMS will do for your workflows down the road. If you’re not quite sure yet, here are a few questions to ask yourself:</p><ol><li>Are you deploying code to make simple content changes (text, images, video, etc.) at least once every other sprint or more?</li><li>If you answered yes to #1, would you like to enable non-engineering teammates to make those content changes to allow engineering resources to focus on other problems?</li><li>Are you sharing content in multiple places that would all need to be adjusted if that content were to change?</li><li>Do your content updates require an approval workflow?</li><li>Do you want to easily support multilingual translations?</li></ol><p>If you said yes to 2 or more of these items (or even if you think you’ll say yes to 2 or more of these items within the next few years), a CMS could be an excellent way to optimize your workflows and free up development resources for new and exciting projects. If you’re at all like me, you’ll look back a year from now with a sense of accomplishment for all of the things you were able to tackle once you streamlined your content management.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8f5899c0d032" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/a-cms-case-study-how-our-implementation-allowed-us-to-scale-for-students-8f5899c0d032">A CMS Case Study: How Our Implementation Allowed Us to Scale for Students</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mission-Driven Product Management]]></title>
            <link>https://medium.com/extra-credit-by-guild/mission-driven-product-management-56a7bf253091?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/56a7bf253091</guid>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[b-corp]]></category>
            <category><![CDATA[guild-education]]></category>
            <category><![CDATA[mission-statement]]></category>
            <dc:creator><![CDATA[Kyle Grant]]></dc:creator>
            <pubDate>Thu, 27 Aug 2020 14:20:20 GMT</pubDate>
            <atom:updated>2020-08-27T14:21:08.376Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*CpKtlcLjHOnwcFWe" /><figcaption>Photo by <a href="https://unsplash.com/@airfocus?utm_source=medium&amp;utm_medium=referral">airfocus</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Almost every company dreams up its own mission statement, but it is rare to work for a company that actually lives by theirs.</p><p>At Guild, our mission is to unlock opportunity for America’s workforce through education. We connect the world’s largest employers and their employees with high-outcome online learning providers. We’ve flipped the traditional education benefit model in ways that alleviate the financial strain on students. Fulfilling <a href="https://bcorporation.net/about-b-corps">our B-Corp status</a> depends on us keeping the best interests of our students in mind.</p><p>Our mission doesn’t talk about disrupting an industry or using a new technology to do something. That’s the difference between legitimately mission-driven companies and other startups who simply have a mission statement: Guild doesn’t succeed unless our students do.</p><p>It feels pretty great to work at a company like this, but it also poses some unique challenges. Here’s my take on how product management is different in an environment like Guild.</p><p><strong>Stakeholders Galore</strong></p><blockquote><em>All stakeholders must participate in the gains and losses of any particular situation.</em></blockquote><blockquote><em>— Christine Lagarde, President of the European Central Bank</em></blockquote><p>Guild sits at the intersection of the world’s largest employers, their employees, and online learning providers. It takes an army of different departments to make that model work.</p><p>Most product managers are used to the struggle of having to coordinate with colleagues across various departments like sales, support, and marketing. At Guild, that challenge is exacerbated by the sheer number of different departments that make this company run.</p><p>With so many stakeholders, it’s a constant challenge to make sure we’re solving the right problems and doing what’s best for our students. Each department has its own unique challenges. This means we’re being pulled in many, sometimes competing, directions.</p><p>It also means we’re going to miss people in our internal communications about product changes. We can try to over-communicate, but then you have to remember that our stakeholders are also receiving a pile of communications from all of the other departments. If you spam people with emails all day, they start to ignore you. This balancing act has taught all Guilders the value of discretion and succinct communication, and it’s something we’re continuously trying to improve.</p><p><strong>We Can’t Solve Everything</strong></p><blockquote><em>The essence of strategy is choosing what not to do.</em></blockquote><blockquote><em>— Michael Porter, Harvard Business Review</em></blockquote><p>This is probably the toughest part of joining a mission-driven organization. We all want to solve all the world’s problems, but we have finite time and resources. Ensuring that we’re working on the most important problems at all times means saying “no” to problems that could be impacting someone’s access to a better life.</p><p>That’s a lot of weight to bear. For example, which is more important: getting more students into programs that will improve their lives or ensuring that those already enrolled in the programs are succeeding? They’re both extremely important, but prioritization is necessary. While we’re fortunate enough to have teams dedicated to both of these challenges, we’re dealing with mini versions of this dilemma on a daily basis.</p><p>It’s hard not to take these problems home with you. Don’t get me wrong: the wins at Guild are great. Knowing we’re helping people makes the bad days good and the good days great. It’s important to keep these wins in mind when we’re making the tough decisions.</p><p>The other issue is that technology just can’t solve some of the problems we’re dealing with. Sometimes the data from partners isn’t there, or we’re trying something that no other company has ever done. Having to deal with limitations when peoples’ livelihoods are at stake is even more soul-crushing than having to say “we’ll get to it later”. “Soul-crushing” might be a bit dramatic, but it’s not easy.</p><p><strong>Fewer Shiny Things, More Solving Real Problems</strong></p><p>That last one was heavy, so I’m going to finish on a positive note. In my experience, Guild is a great place to be a product manager. We have a team of leaders who are experienced in avoiding common product management pitfalls. Having an organization centered around an inspiring mission doesn’t hurt either. A lot of our common language is centered around helping students, rather than making money.</p><p>Product organizations, especially young ones, often run into a trap of “future selling” or <a href="https://www.mironov.com/sales-led/">sales-led developmen</a>t. In other words, they sell their product on the possibilities of what’s coming down the pipeline, rather than selling the product on the merits of its current value proposition. This leads to a lot of building features that don’t solve actual user problems but make for great sales pitches. Product managers loathe this type of thinking.</p><p>We try to avoid sales-led development at Guild, but we also avoid guiding key metrics that are sales-related. In fact, our yearly company-wide goal isn’t a revenue number like it might be at most companies; it’s the total number of active students enrolled in a program. This is done intentionally so that we don’t guide students toward a program just because it brings in more revenue or has a higher margin. Our annual goal is a product of our B Company status and our mission to focus on student outcomes.</p><p>This all means that we’re not a <a href="https://cutle.fish/blog/12-signs-youre-working-in-a-feature-factory">feature factory</a>. We aren’t churning out features as quickly as possible without measuring the results. I’ve been told on multiple occasions to focus on the problem, rather than the product. These are signs of a healthy product organization. I feel lucky to work in a place that gives us the freedom to solve real problems and to do it the right way.</p><p>Product management in a mission-driven environment is both rewarding and challenging. I’m grateful for the opportunity to experience both. Once you change your mentality from “how do we make more money?” to “how do we help more people?” it’s hard to imagine going back.</p><p>My suggestion for anyone looking for a product job, starting their own business, or looking to motivate their staff is to find a mission that inspires. Build a team of missionaries around it and go all in. Look into becoming a B-Corp. At the end of the day, if you’re going to be devoting your blood, sweat, and tears to something, you may as well make a difference while doing it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=56a7bf253091" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/mission-driven-product-management-56a7bf253091">Mission-Driven Product Management</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The expected (and unexpected) rewards of building a data culture at Guild Education]]></title>
            <link>https://medium.com/extra-credit-by-guild/the-expected-and-unexpected-rewards-of-building-a-data-culture-at-guild-education-1639539dde7?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/1639539dde7</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[looker]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[guild-education]]></category>
            <category><![CDATA[business-intelligence]]></category>
            <dc:creator><![CDATA[Sean Mckeever]]></dc:creator>
            <pubDate>Thu, 20 Aug 2020 14:02:15 GMT</pubDate>
            <atom:updated>2020-08-20T14:10:27.890Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YrgTAUYuzBxsBnZJ-5hCEA.jpeg" /><figcaption>Art by Erin Rhodes</figcaption></figure><h3>The Expected (and Unexpected) Rewards of Building a Data Culture at Guild Education</h3><p>At Guild Education, our goal is to put frontline workers back to school, debt-free, through employer-funded tuition assistance. Within the platform, our student coaches personalize the education experience by helping workers map out their program and providing them with ongoing support from start to finish. This process ends up gathering a LOT of data — data from our partners, data from schools, data from our website, and on and on. The trick is, this data is only as good as the insights and utility that are derived from it.</p><p>The term “data culture” refers to a company culture that empowers stakeholders at whatever level to explore data to learn and inform their decisions. In order for this to take place, not only does data and reporting need to be easily accessible, but end users need to be informed and educated about best practices and availability of certain data. As we began our journey of creating a data culture here at Guild, our coaches were at the forefront, helping us build a culture where everyone can feel empowered to do impactful work with data.</p><h3>Kicking things off</h3><p>To begin building a data culture at Guild, the team started by implementing Looker as our <a href="https://looker.com/platform/overview">data platform</a> and by hiring a data analyst and two business intelligence (BI) specialists, including me.</p><p>When the three of us joined Guild Education, we were faced with thousands of spreadsheets, no single source of truth, and a culture that frequently leaned on gut-level decision making as a result. Our mission as a data team was to roll out Looker to the company by transitioning a few departments at a time until the whole organization was plugged into our new way of approaching data analytics.</p><p>Fast forward to today, the successes we’ve experienced solidify that we are well on our way to becoming a fully <a href="https://looker.com/blog/data-culture-2020">data-driven company</a>. Roughly 50–60% of the company logs into Looker on a weekly basis — which is truly fantastic considering we’ve more than quadrupled the number of employees from 150 to 700 during the last two years since adopting Looker. In addition, data-centric roles at Guild Education have grown considerably, with about 30 people now working under the umbrella of analytics, BI, research, and operational enablement.</p><h3>Transforming our Student Success program with Looker</h3><p>One of the programs that’s completely transformed with the help of data and Looker — which we are especially proud of — is our Student Success program.</p><p>In the Student Success program, coaches work directly with students in the program and provide ongoing support throughout their entire educational journeys.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*RkDAnfEE44G0t-GM.png" /></figure><p>When the program first launched, coaches were able to remember their students by name off the top of their head; but as enrollments increased, even spreadsheets were not able to accurately keep track of caseloads. We needed a singular source of truth that could be updated in almost real time.</p><p>As a solution, we created what we call “Student Rosters,” which leverage Looker dashboards and serve as operational tools to manage a coach’s student outreach. When we rolled out the Student Rosters to the first group of coaches, they were thrilled about how streamlined the process of reaching out to students was with the new tool. As we continued the rollout, however, things started to get complicated.</p><h3>Growing pains</h3><p>The Student Success team was growing at a wild rate — climbing from about 20 to 90 employees to date. To keep up with growth, coaching teams started developing subteams, many of whom were using data from different sources and with different outreach strategies, making consistency and accuracy a major issue. Plus, because programs and benefits changed over time, the roster dashboard had to be modified frequently.</p><p>I was the main point of contact for all the coaches. Before long, my days were increasingly spent fielding questions, training users, and making adjustments to the dashboard. I had to figure how to get my time back under control.</p><h3>From users to data-empowered task forces</h3><p>Recognizing these complications with the Student Rosters as a bottleneck for making progress in other areas at Guild, my team and I brainstormed ways to empower more users with data in the midst of our rapid growth. This led to the creation of our ambassador groups — first the Student Success BI task force, and 6 months later, the Education Coach BI task force.</p><p>For the creation of the first group, we partnered with Student Success managers to find one coach for each sub-team in the department. The managers quickly bought into the goal of developing a task force given they were well aware of the need for data experts throughout the department.</p><p>Every two weeks, I’d spend an hour with the group of coaches, getting updates about their world, training them on the Looker platform (and analytics more generally), and discussing all of our projects in the works.</p><p>Through establishing the BI task force, the coaches and I were able to get a finger on the pulse of what was happening across the organization.</p><h3>A data-driven evolution</h3><p>These ambassadors became the go-to leaders for fielding user questions, training coaching teams, and for creating updates and documentation for ongoing Looker work. For people like me on the BI and analytics team, the program opened our eyes to the unique problems each group faced and helped us stay up to date on the growth of the coaching program.</p><p>Unexpectedly, the Student Success task force also became an engine for new BI work. They blew my highest expectations out of the water and started owning virtually the entire process: building requirements, prototyping, testing, and deploying new dashboards and panels — with hardly any help from the BI team. They became so self-sufficient that they ended up fully rebuilding all four student rosters and significantly improved their outreach strategy in the process.</p><h3>Accelerating opportunities for professional development</h3><p>The task forces accomplished our original goal of creating a team of experts beautifully, but that wasn’t all. They also had a tremendous impact on the professional development of the people involved.</p><p>A great example of this is the career evolution of one of our Student Success coaches. From the moment Alex joined the task force, she led the charge on the ideation and creation of new dashboards and metrics and acted as a facilitator for the other members of the team. She proved herself to be tremendously hardworking and gifted in her work with both people and data. Six months after she joined the first BI taskforce, we officially hired her onto the BI team. Ultimately, I handed over leadership of the team to her, leaving the group of coaches to be led by a former coach!</p><p>Today, five out of 13 members from the first two iterations of the Student Success task force now hold new and more data-intensive jobs at Guild, with many of the remaining coaches also in new positions following promotions. And each of the eight coaches from the first iteration of the Education Coach task force ended their six-month involvement in the program with new jobs within the company: two were promoted to team leads, two were promoted to managers, and the remaining four were hired into other teams.</p><p>Considering all of these evolutions, both expected and unexpected, the main learning is that if your company is on its way to becoming <a href="https://info.looker.com/webinars/how-to-build-a-data-driven-company-from-infrastructure-to-insights-2">truly data driven</a>, providing opportunities for people to build their Looker chops will only make them more valuable as an employee. It’s a huge advantage to be able to pull reports and make data-driven decisions, and having that knowledge gives everyone increased job mobility.</p><h3>Tips for building a data culture</h3><p>As we continue to build upon the practices that create our data culture here at Guild, I’ve had the chance to reflect on why this initiative has been so successful. Here are some of my top tips for other organizations looking to kickstart their own data culture:</p><h3>Make data seem approachable and fun (because it is!)</h3><p>In general, most people are apprehensive about data. Many associate data with messy spreadsheets, math, and/or computer programming. It’s important to teach early on that analytics is more about the questions and thought process than it is about the actual work with data. Can you ask questions and brainstorm different approaches to answering those questions? That’s analytics!</p><p>One thing that helps a lot when teaching folks about analytics is not to take yourself too seriously. The goal isn’t to be the expert on anything; the goal is to encourage others to explore the platform and for them to become the expert. The less approachable you seem as the facilitator, the less approachable learning Looker will seem. If you talk openly about your previous data blunders or about the imposter syndrome many of us have experienced, others might start thinking, “Hey, I have been feeling like this wasn’t for me, but this guy seems to know his stuff and he has felt that same way before. Maybe I can learn this.”</p><h3>Form small groups and use the right collaboration tools</h3><p>When it comes to creating groups for learning, I find it easier and more productive to hold meetings with no more than 10 coaches and 1–3 analytics team members for support. At Guild, we have bi-weekly meetings in the calendar and also use a Slack channel to keep the communication flowing.</p><h3>Follow a standard structure with each task force</h3><ul><li>Provide an overview: At the first meeting of a new task force group, I do a rundown on the purpose of the group and provide an overview of the world of data and BI, linking this back to the business. We then spend the second half hour of that meeting talking about the different types of data the team can find on Looker at the dashboard level, and then do a quick introduction to Explore mode.</li><li>Take a deeper dive: The second meeting is focused on Explore mode, so that everyone can understand how to experiment with ad-hoc queries or perform one-off calculations. In this meeting, coaches learn about the differences between dimensions and measures, how to save Looks, and more. At the end of the class, I ask them to build a specific panel, and we then review the panel as a group. I have found that it takes about 3–5 meetings (plus some project work between meetings) to really get the whole team comfortable using Explore mode.</li><li>Incorporate training into every meeting: After the first two meetings, we generally spend about 20 to 40 minutes on training and hands-on exercises with Looker. I find that consistent training up front reaps big dividends down the line. Eventually, we wind down to only 10 to 15 minutes of training and 45 to 50 minutes discussing the issues specific to each group.</li><li>Exchange ideas and set goals: At the beginning of every meeting, we share everything we’ve accomplished during the previous two weeks, and, at the end of every meeting, we decide what we want to work on over the next two weeks. Think of it as an informal sprint planning session.</li></ul><h3>Bring in new data enthusiasts</h3><p>Every four to six months, we rotate out some members of the group to get some fresh faces. We want to build a diverse group of power users, so we like to spread the love and bring new folks into the program so that we end up with as many Looker experts as possible.</p><p>As folks are trained and then move to other teams, the number of people with in-depth Looker knowledge grows. Before you know it, you’ll have experts across your whole company.</p><p>In other words, you’ll have a data culture.</p><p><em>This piece was originally published by </em><a href="https://looker.com/learn/recorded-demo?utm_campaign=701E0000000ocUe&amp;utm_ad_group=looker_exact&amp;utm_content=looker_exact&amp;utm_ppccampaign=na_search_brand&amp;_bt=213890334185&amp;_bk=looker&amp;_bm=e&amp;_bn=g&amp;gclid=EAIaIQobChMItvub6fqp6wIVyyCtBh0ZRguPEAAYASAAEgIaL_D_BwE"><em>Looker</em></a><em> on August 3, 2020. Click </em><a href="https://looker.com/blog/building-a-data-culture-at-guild-education"><em>here</em></a><em> to read the original post.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1639539dde7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/the-expected-and-unexpected-rewards-of-building-a-data-culture-at-guild-education-1639539dde7">The expected (and unexpected) rewards of building a data culture at Guild Education</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Information Flow in Three Work Cultures]]></title>
            <link>https://medium.com/extra-credit-by-guild/information-flow-in-three-work-cultures-fe6c30d26bab?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/fe6c30d26bab</guid>
            <category><![CDATA[guild-education]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[company-culture]]></category>
            <category><![CDATA[qa]]></category>
            <dc:creator><![CDATA[Quentin Hartman]]></dc:creator>
            <pubDate>Thu, 13 Aug 2020 15:51:24 GMT</pubDate>
            <atom:updated>2020-08-13T15:54:39.475Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OUwkL1b-1kr17-Kzt56Stg.png" /><figcaption>Art by Erin Rhodes. Four employees work together to assemble the pieces of a large puzzle.</figcaption></figure><p>Nothing defines a group of people more than its culture. More than anything else, culture shapes what it feels like to be a part of that group. “Culture” is an expansive concept, and can include everything from the most foundational beliefs and major ceremonies of a group, to the most mundane little details of daily life. Art, music, food, speech patterns, clothing, architecture… the list could go on. But what does culture mean in a business context? What is your culture at work? How does the culture <em>at</em> work affect your ability<em> to</em> work?</p><p>Work cultures are distinct from other cultures in that they are much more narrowly focused, and are often created with a lot of intention. If culture in general describes how a group lives, culture in business describes how a group works. A lot of scholarly effort has been put into studying and defining work culture. One of the most useful models of organizational culture was put forth by Ron Westrum in his 2004 paper “<a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1765804/pdf/v013p0ii22.pdf">A Typology of Organisational Cultures</a>,” where he defines the three cultures model. The cultures are <strong>pathological</strong>,<strong> bureaucratic</strong>, and<strong> generative</strong>. Westrum found that one of the most telling attributes about the dominant culture is how the people and processes within an organization handle information. Especially telling is the handling of information about failures, flaws, weaknesses, or other shortcomings. Let’s call this Toxic Information.</p><p>A pathological culture is as negative as it sounds. This culture is focused primarily on the acquisition and retention of power by individuals within the organization. One of the primary means to maintain this power is to control the flow of information, especially Toxic Information. In a pathological culture, Toxic Information is frequently weaponized. It is used against others in the organization to ensure the person who has the information maintains their power.</p><p>It’s also common in a pathological culture for information to be suppressed. When it can’t be suppressed, a scapegoat is found to “take the fall” for the organization. This results in a large number of other negative behaviors, including low cooperation, blame, siloing information, hiding failures, and fear of innovation. These traits are particularly dangerous for creative work like software development. They encourage people to always choose the “safe” paths, which hinders innovation. They create adversarial social patterns between teams, which leads to infighting, and known problems getting ignored until they cause issues in production.</p><p>Strongly pathological organizations can be effective for short periods of time so long as they pay very well or have some other amazing upside for the individuals, but people tend to become unhappy and burn out or become disengaged quickly. As a result, these organizations tend to have <a href="https://www.businessnewsdaily.com/8718-attracting-retaining-strategies.html">high turnover rates</a>, and there is a strong <a href="https://acgc.cipe.org/business-of-integrity-blog/the-characteristics-of-corrupt-corporate-cultures/">correlation with markers for corruption</a>. In fact, organizations that go too far down this road often collapse because they are held together only by individual self-interest, and fundamental problems are rarely addressed.</p><p>A Bureaucratic culture, on the other hand, focuses primarily on rules. Almost every large organization has some element of bureaucracy in its culture. Virtually any time you see a standard form, or have to look up a process, you’re seeing bureaucracy in action. Some amount of it is necessary, even healthy, for organizations to scale efficiently. However, there are pitfalls. Bureaucracy can often lead to “not my problem” attitudes because areas of responsibility are very narrowly defined. Communication between groups is often limited, and sometimes discouraged. This can be because an “us vs them” attitude has arisen because people have begun to identify with their immediate coworkers, but not the organization as a whole. It can also be because the “proper channels” for that communication are too onerous. The novel or interesting tend to be actively discouraged because change of any kind is seen as too risky. Individuals are disempowered to make decisions unless they are formally part of the leadership group. All this leads to Toxic Information being ignored or shuffled from place to place, allowing problems to go unaddressed for extended periods of time.</p><p>Bureaucracies also tend to be strongly justice seeking. When Toxic Information does finally come to light, they will not generally try to suppress it or dodge responsibility, but they will be quick to find some form of punishment or restitution to execute “justice.” This leads to a feeling that the issue is “dealt with,” but will often fail to address the root problem. Perhaps worse, bureaucratic thinking will cause people to be overly reliant on “proper channels” in a crisis. Westrum points to the communication failures between police and firefighters responding to the September 11th attacks as a catastrophic example of this force in action.</p><p>Bureaucracies can be very effective, however. A great example of this is a military organization. Most world class militaries are very effective at meeting their objectives, but most of the people who are there aren’t performing creative work, and relatively few of them are likely to be happy enough to stay there long term. Their biggest weakness, though, is their inability to adapt and act efficiently. The often onerous process to get changes approved takes time and effort away from actually implementing those changes. This is not a good environment for developing software, where the pace of change is constantly accelerating.</p><p>Finally, generative cultures are fundamentally mission and performance focused. The high-level goals of the organization are clearly articulated, and individuals are empowered within their spheres to contribute to that mission. Cooperation and collaboration are encouraged. Novel ideas and experiments that are likely paths to better outcomes are encouraged. Most importantly, Toxic Information isn’t seen as toxic at all: it’s an opportunity for improvement. As a result, strongly generative organizations will actively seek out Toxic Information as a means of self-improvement. When Toxic Information surfaces, they will look for contributing factors and work for true fixes to underlying problems. Justice seeking, scapegoating, and suppression should all be very rare, and ideally, entirely absent.</p><p>People in generative cultures tend to be happier on average than in other cultures because they can clearly connect their work to the higher goals of the organization and are empowered to make a difference. This clearly is the best environment for creative work like software development, but it’s not without its risks. The biggest challenge in a generative culture is scaling it to large numbers of people. As people know each other less, trust is more difficult to maintain, communication is less frequent, and efficiency can be lost as people lose track of what others are doing and overlapping work starts happening.</p><p>At Guild we’re working towards a strongly generative culture, even in the face of our rapid growth. Not only does this cultural model provide for the healthiest and happiest environment for the people that inhabit it, but there are also strong correlations between generative attributes and effective organizations. We’ve already established that the handling of Toxic Information is a strong indicator for the kind of culture that exists in an organization, but it goes beyond that. The way that all information flows is important. In a generative culture there is a strong emphasis on getting the right information to the right people at the right time. That is a challenge at the best of times, and in an organization that is growing as rapidly as Guild is, that challenge is magnified.</p><p>At the highest level, Guild is meeting that challenge with a focus on transparency and candor. Frequent communications about the state of the business, progress on initiatives, and upcoming challenges are all topics one can expect to see. But several less obvious efforts within the technology organization also promote a generative culture.</p><p>The establishment of the QA, Security, and DevOps teams this year are prime examples of organizational changes that have deep cultural importance. On the surface, these teams are centers of specialized skills and considerations that are fundamental parts of building and delivering software, but they are also key actors in facilitating information flow and cooperation across teams.</p><p>DevOps is most directly and obviously engaged in helping information move around and fostering cooperation. It’s right in our team mission statement. We spend a large portion of our time interacting with developers from all over the company, supporting their work around automation, observability, and infrastructure. We bring our own skills to the table, helping squads make stronger and more consistent decisions, more quickly. Our goal here is to be a super responsive first line of support to get developers unblocked. But we would be making a big mistake to think we have all the answers. We are also collaborating with the individual dev squads. We are collecting good ideas that we find in one squad and sharing them where they might be useful on another. Sometimes, we even integrate them into our core patterns. The observability efforts that we are shepherding are entirely about surfacing the right information to the right people at the right time.</p><p>QA’s role here is also pretty obvious. I know the joke is that all QA does is find creative ways to break things, but really, it’s about making developers aware of ways the software we create might not do what we expect. Ideally, it also creates that awareness as quickly and easily as possible using sophisticated testing and automation tools. The benefit of focus here is twofold: we have better awareness of flaws, and we can spend less total effort finding them. Once the problems are found, QA also makes sure that loops get closed with the project managers and devs who own that code so that decisions can be made about how and when to address the issues.</p><p>Security helps information flow through auditing and compliance efforts. The relationship here isn’t obvious, but it becomes apparent if you dig below the surface a bit. Audits are fundamentally about discovering where we can be better, and compliance compares us to recognized standards for good processes and practices. When we’re proactive about compliance and can show evidence that we meet these standards when partners ask for it, we protect ourselves from disruptive audits being imposed on us and we make it easier to get contracts for new deals signed. Our security team is also engaged in this effort through a formal Application Security program that is being spun up. At its core, the program aims to ensure that developers know about secure coding best practices and that they are warned about possible vulnerabilities in their code before it makes it into production.</p><p>It takes a lot of work to create an organization with a great culture. There are countless aspects to be considered, but perhaps the most important one is how the organization handles information, especially information we’ve called Toxic Information. Is it hoarded to be used as a weapon for personal gain? Is it hindered by bureaucratic roadblocks? Or, does information flow freely to where it’s needed when it’s needed? The latter is what we’re aiming for at Guild. It’s a major component of a generative culture, where people are happy and the organization is able to meet the high-level goals at the heart of its mission. Information flow is only one part of what it takes to get there, though. What else can you do in your role, with your teammates, to get your organization closer to that strongly generative ideal?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fe6c30d26bab" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/information-flow-in-three-work-cultures-fe6c30d26bab">Information Flow in Three Work Cultures</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mentorship via Debugging: How to Turn a Bug Into A Learning Experience]]></title>
            <link>https://medium.com/extra-credit-by-guild/mentorship-via-debugging-how-to-turn-a-bug-into-a-learning-experience-9c86fc8f44c?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/9c86fc8f44c</guid>
            <category><![CDATA[mentorship]]></category>
            <category><![CDATA[guild-education]]></category>
            <category><![CDATA[extra-credits]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[debugging]]></category>
            <dc:creator><![CDATA[Tyler Long]]></dc:creator>
            <pubDate>Thu, 06 Aug 2020 13:58:55 GMT</pubDate>
            <atom:updated>2020-08-06T13:58:55.666Z</atom:updated>
            <content:encoded><![CDATA[<h3>Mentorship via Debugging: How to Turn a Code Bug Into A Learning Experience</h3><p><strong><em>⛰ 🐛 🛫 🛬 🐛 🏖</em></strong></p><p>Here at Guild, we like to foster a welcoming and inclusive engineering environment. In this series we will explore some ways to be an effective mentor as well as an effective mentee. For this segment, let’s use bug hunting as a tool for mentorship! Bug hunting is an invaluable skill, and it becomes an even greater challenge when you are dealing with distributed systems.</p><p>Bugs are a part of all systems, and tracking them down is a part of our jobs as developers. But when a bug is no longer a single fly buzzing around a program and becomes a fly stuck in a web of systems, debuggability is more cumbersome and tedious. In order to successfully hunt it down, we need to enhance our understanding of what it means to trace bugs through systems.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qrzFtc7fo7kTt50t" /><figcaption>Photo by <a href="https://unsplash.com/@guillaumedegermain?utm_source=medium&amp;utm_medium=referral">Guillaume de Germain</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>A systems model is an essential tool for tracing complex problems. Being able to visualize, either mentally or with a diagram, provides a “big picture” context that allows you to identify spaces where bugs can hide. Teaching newer engineers how to trace bugs is a great way to build out mental models for your systems and to level up people on your team. Let’s think of an example.</p><p>Imagine that a user reports that they are having trouble logging in. A likely instinct is to check the authentication code. This a logical first assumption, as that code is directly responsible for the actions taken by the user in order to access your system. In a monolithic architecture, we could trace the call stack through one system; but when services are abstracted and microservices adopted, we need a mental model in order to properly map out what may be going wrong. We need to know how the code we <em>think</em> is responsible for the bug is interacting and affecting other code and systems around it.</p><p>Let’s use a hypothetical authentication flow in which we have many applications, as well as many services driving the data and logic behind these applications.</p><p>At the highest level, this happens:</p><p><em>User visits a login page, enters credentials, and authenticates into </em><strong><em>Application A</em></strong><em>.</em></p><p>If we dig slightly deeper, we understand that the process looks more like this:</p><ol><li><em>Code is run to validate that the user exists in our database, passwords match, and the user has all the data we need to establish a connection into Application A.</em></li><li><em>The validation returns true, so grant the user a JSON Web Token (JWT, pronounced “JOT”), establish a session, and carry on with your day.</em></li></ol><p>But thanks to our mental model, we know that reality looks much closer to this.</p><ol><li><em>User visits a login page, enters credentials, and authenticates into Application A (So we have a Single Page App that renders the login form, and handles all client-side validations. We can call this </em><strong><em>Service One</em></strong><em>)</em></li><li><em>Service One posts data to a backend service to look up the user and validate their credentials (So, we now have another server-side application handling database lookups and validations as well. We call this </em><strong><em>Service Two</em></strong><em>)</em></li><li><em>Service Two does some querying, finds the user, formats the User object into JSON, and passes it back to Service One.</em></li><li><em>Service One receives this data, formats it into a JWT, and hands it over to Application A.</em></li><li><em>But wait! Application A needs some additional data in order to fully render an experience for the user! Application A then queries </em><strong><em>Service Three</em></strong><em>, which finds the missing data and passes it back to Service Two, which then re-formats the JWT, and then the JWT is passed back to Application A. Finally we are good to go!</em></li></ol><p>Phew. That’s a lot of services doing a lot of work for an authentication flow! Without an understanding of how all services interact and work together, it’s much harder to determine the root cause. Sure, we can keep making educated guesses about where to target our search, but with a clear mental model of the system, the task becomes less daunting. If you were part of the initial buildout of this system, or helped in architecting it, finding a bug may be much easier; however, imagine what this feels like as a brand new person on the team, or a junior engineer just starting out… it’s confusing!</p><p>In a way, I think of debugging as a<a href="https://git-scm.com/docs/git-bisect"> git bisect</a> of our system — that is, let’s start with the outlying layers and drill inward from there. With a git-bisect we can use a binary search in order to find the commit that introduced the bug. In a system-bisect we can utilize a binary-like search to find what system contains the bug. The mental concepts are the same; only the mechanisms are slightly different. And having a model of our systems allows us to bisect more effectively.</p><p>Effective bisecting can help get around an age-old truth about engineering: what you may think is obvious is not, in fact, obvious. Yes, sadly, your assumptions after months or years of working in a system can sometimes leave out important and relevant details for unearthing your bug. So, walk through the basics as part of your process, and train teammates to do the same.</p><p>Now, when debugging, context is king. If the engineer debugging this issue has been around for a while, or helped architect some of the system, it is much easier for them to hold on to a mental model while going through the debugging process. Maybe they know that certain users are always missing data Y when they come from a certain population — in fact they have that ticket lingering in the icebox ready to fix this exact issue! So I suggest two things here:</p><ol><li>Diagram <em>and</em> document everything. I know it may sound redundant, but strong documentation can come in two forms: written and visual. Use both, please! Some people are visual learners, and having the design of your systems in this format is extremely helpful in turning everyone on your team into effective debuggers. Others will prefer to read about the design of the system, so having written documentation is equally valuable. Some folks, like myself, prefer both. It is nice to be able to visualize a system with an architecture diagram, while at the same time having solid documentation around each available service. By visualizing the flow, I know that Service One makes a POST request to Service Two. But what endpoints are available on Service Two? How is authentication handled on that backend? And what gets returned if something is amiss? These are the types of things that documentation and diagrams can help with! So please, write good documentation, but also help other people who learn differently with visual diagrams. When you are the only person who understands a system fully, and you are out of the country for two weeks on vacation, your past self will thank you.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*L9c-Do3QUVx5ZTqa" /><figcaption>An example of visual documentation for the authentication flow we described above.</figcaption></figure><p>2. Pair with people! If you are the resident expert on a particular system, and a hairy bug comes in or a system is down, can you grab a more junior person on your team and have them ride along as you debug? This works twofold:</p><p>a) You are sharing a lot of knowledge with your team and showing someone who doesn’t have as much experience how <em>you</em> debug. We all do this slightly differently, and there are a lot of techniques that can be shared! Maybe you love networking, but your more junior teammate doesn’t know much about it, so show them! They will get stronger and you will grow to understand them more.</p><p>b) You are jaded. Yeah, you, the super senior. Other people on the team, especially less experienced or newer members, have good questions and fresh eyes. They see things that you overlook, and they can give you a different vantage point. Pairing on bug triaging is a valuable exercise for you and your teammates. Help them grow and share the knowledge; sure, maybe this isn’t always feasible, but if some non-mission critical bug is popping up, use it as a mentoring opportunity and level up your friends. Again, when you are on vacation, your past self will thank you.</p><p>Not sure where exactly to start? Here’s a list of the common places I check when I’m debugging:</p><ul><li>Did a bad deploy cause this? Walk new teammates through how to find deploys and versions, how to roll back bad deploys when they happen, and how to do a postmortem or retro on things when they go wrong.</li><li>Did the bug reporter send a screenshot? Can you see what browser they are using? Is there a chance that the problem is browser specific? Does the screenshot show what type of error they received? Little bits of knowledge that feel unimportant or obvious could turn out to be relevant, and they might not be apparent to someone newer to your system. For mentoring, my motto is: Ask, don’t assume. Asking clarifying questions helps you, the debugger, clarify the problem statement for yourself while also clarifying things for your mentee.</li><li>Where are your logs? You have logs, right? Logs can be noisy and hard to trace. Show your teammates the little tips and tricks you’ve picked up along the way. If there is a log aggregator — Splunk, Datadog, Cloudwatch, whatever — show teammates how to query it effectively. If your logging is set up to tag events with a piece of unique data, find that data and show new developers how to format those queries to get the most value out of your logs.</li><li>What is handling your networking? Do you have proxies and DNS and workers set up? Some engineers are not as familiar with networking and may not realize that it could be an issue. Show them your networking code, where it lives, what it does, etc. Again, what you think is obvious isn’t always crystal clear for others, or even for yourself!</li><li>Is something down? No, seriously, is AWS having issues in us-west-2? Is a third party tool having issues? Sometimes the simpler the issue, the more easily it can be overlooked. Sanity check your third party systems!</li><li>Reproduce the bug. Spin it up locally and try to reproduce. Get a runtime debugger involved. Are you using Rails?<a href="https://github.com/pry/pry"> Pry</a> can be your best friend. Are you using React?<a href="https://chrome.google.com/webstore/detail/react-developer-tools/fmkadmapgofadopljbjfkapdkoienihi?hl=en"> React Dev Tools</a> and good old fashioned `console.log` can be life savers. By reproducing the bug locally, we bring the logs to our machines and have strong tools to drill into each bit of code and find the weak link.</li></ul><p>So, there you have it — using debugging as a means of mentoring. Though not hard and fast, I think some of these practices and ideas can really help other engineers on your team get a better grasp of your systems, while also providing immediate value to users. Now, a junior engineer just joined your team, and in the first week you’ve shown them the lay of the land while also fixing a bug affecting users. That’s a win-win scenario that really helps empower your teammates. It shows them that, yes, we all write bugs; and yes, you really can do this!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c86fc8f44c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/mentorship-via-debugging-how-to-turn-a-bug-into-a-learning-experience-9c86fc8f44c">Mentorship via Debugging: How to Turn a Bug Into A Learning Experience</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Creating a Working from Home Environment]]></title>
            <link>https://medium.com/extra-credit-by-guild/creating-a-working-from-home-environment-57d87f556a54?source=rss----50d55bb8b267---4</link>
            <guid isPermaLink="false">https://medium.com/p/57d87f556a54</guid>
            <category><![CDATA[guild-education]]></category>
            <category><![CDATA[remote-work]]></category>
            <category><![CDATA[work-from-home]]></category>
            <category><![CDATA[pandemic]]></category>
            <dc:creator><![CDATA[Sean O'Keefe]]></dc:creator>
            <pubDate>Thu, 30 Jul 2020 14:35:19 GMT</pubDate>
            <atom:updated>2020-07-30T14:42:55.439Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>Creating a Working Environment: Part II</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*db3BstvB0wUCrky0ylIxCw.png" /><figcaption>Art by Erin Rhodes</figcaption></figure><p>I first wrote about <a href="https://medium.com/extra-credit-by-guild/creating-a-working-environment-48c081ae74c4">creating a working environment</a> to make a point about malleable plans to help approach work. Despite coming from a development perspective, those strategies can be applied broadly. I wrote about creating spaces that work for you — adding to and refining those tools and techniques as you discover which ones are useful.</p><p>I centered that post around spaces — physical spaces, virtual spaces and mental spaces — with the primary idea that these all interact and build upon each other. The main idea was to create a foundation; to keep simple things simple, so that you have the space, energy and focus to tackle harder challenges. Harder challenges are where you want to spend most of your time, and they are way more fun.</p><p>We all have foundations of knowledge and memory that take different shapes as they are formed by our individual experiences. Whenever I revisit core assumptions that rely on incomplete mental models or lack information, I discover not cracks in my own foundations but new areas to explore and improve. That introspection allows me to move forward with a better understanding and also strengthens my ability to articulate alternatives and trade-offs. I improve my technical skill, communication and ability to collaborate with others by incorporating other ideas and approaches.</p><h3><strong>Routine is important, but not the same for everyone</strong></h3><p>Foundations that build on each other and allow continued growth are important, but with the previous post I had left out a crucial element: all of our co-workers.</p><p>Working from home is new to a lot of people. Even for those with experience working remotely, the current pandemic is something altogether different. Remote work used to be more of a solitary experience. Now, it is an “alone together” kind of situation. The future of work is unknown, but it is becoming obvious that it will change faster than anyone anticipated.</p><p>If it is a struggle to adapt, changing up your mental model will likely be the first step. What was thought of as temporary is becoming more… if not permanent, then certainly indicative of a lasting impact. It is likely that you have developed habits and routines and figured out a way to make it all work. Are some of those habits helpful and others less so? Maybe you’ve used Slack to stay connected to your team on the one hand, but on the other you realize you sit for hours and hours most days without a break and can’t feel your legs. Did your routine happen with intent, or did you sort of just fall into it? There are no right answers here, but plenty of good suggestions (that’s called foreshadowing).</p><h3><strong>Blending routines</strong></h3><p>It is one thing to figure out how to organize working spaces, schedules, and routines for yourself. It is another to figure out how to blend everything with everyone you are working with, both for your immediate team and the organization as a whole.</p><p>Communication is key, and while this is always true, there is less wiggle room when everyone is remote. Rather than a bug, this can be a feature. It is easy to make assumptions about what information has been communicated when everyone is in an office; it is necessary to be more explicit when there are remote offices and workers.</p><p>The trick is to find what is working well for you and those you work with. Double-down on the things that help, and don’t get stuck in the mire with the things you are “supposed” to do that aren’t actually effective.</p><p>In that spirit (and payoff of the prior foreshadowing), I asked other Guilders to share suggestions that have worked for them as we’ve all learned to work together apart:</p><h3><strong>Physical Space</strong></h3><blockquote>If you have the space for it, it helped me 10x to have my work space in a completely different room from the chill spaces in my home. Makes it more like you’re actually “commuting” and switching mindsets when you sit down to work.<br> <em>— </em><a href="https://www.linkedin.com/in/kdubuisson/"><em>Kate Rose</em></a></blockquote><blockquote>Having a dedicated space where I do nothing but work really helps keep me focused.<br> <em>—</em> <a href="https://www.linkedin.com/in/ryan-shanley-0220a27/"><em>Ryan Shanley</em></a></blockquote><blockquote>Low amount of audio or visual distraction. Regularity in location is useful for me. <em>— </em><a href="https://www.linkedin.com/in/josh-mattila-0828b05/"><em>Josh Mattila</em></a></blockquote><blockquote>Per sleep hygiene recommendations, never work in the room you sleep in.</blockquote><blockquote>Noise isolation. Office space or home, sometimes the ability to block out noise is critical for focus.<br><em> — Derek Koloditch</em></blockquote><blockquote>I have a hard rule of never working on the couch. Mainly because it’s a slippery slope from ‘sitting on the couch working’ to ‘reclining on couch with blanket and snuggly pillow with tv on in background “working”’<br><em> — </em><a href="https://www.linkedin.com/in/jalena-taylor/"><em>Jalena Taylor</em></a></blockquote><h3><strong>Mental Space</strong></h3><blockquote>I have been taking lunch-time walks which I have really been enjoying<br><em> — </em><a href="https://www.linkedin.com/in/andylushman/"><em>Andy Lushman</em></a></blockquote><blockquote>Going outside or somewhere other than your home desk to eat lunch<br><em> — </em><a href="https://www.linkedin.com/in/erinkrhodes/"><em>Erin Rhodes</em></a></blockquote><blockquote>I’ve been going for a bicycle ride to get outside. Great way to get exercise and detach the brain from work for a bit.<br><em> — </em><a href="https://www.linkedin.com/in/nickashirley/"><em>Nick Shirley</em></a></blockquote><blockquote>Having a standard morning routine (that does not include falling out of bed/into desk chair to start work) worked wonders for my mental happiness. I didn’t realize how much the ceremony of commuting to the office prepared me for the day until I didn’t have that anymore.<br><em> — </em><a href="https://www.linkedin.com/in/jalena-taylor/"><em>Jalena Taylor</em></a></blockquote><blockquote>Take a lunch break!! +100 to that. It’s easy to get into the habit of just being <strong>always</strong> on when you’re WFH so I always physically step away and take a lunch out on my patio.</blockquote><blockquote>I would also recommend the Pomodoro method or similar to give yourself mental breaks throughout the day.<br><em> — </em><a href="https://www.linkedin.com/in/kdubuisson/"><em>Kate Rose</em></a></blockquote><blockquote>Give yourself a break. This is not about making shakes, but now I kind of want a shake… vanilla ice cream, splash of whiskey, splash of the cherry syrup that you have left over after using all the maraschino cherries to make manhattans…<br><em> — </em><a href="https://www.linkedin.com/in/sean-okeefe/"><em>Sean O’Keefe</em></a></blockquote><h3><strong>Time Management</strong></h3><blockquote>I’ve found using a calendar management app like Clockwise to be really helpful with proactively blocking out focus time on my calendar<br><em> — </em><a href="https://www.linkedin.com/in/travis-haby-70ab0489/"><em>Travis Haby</em></a></blockquote><blockquote>I added a daily alarm on my phone at 5:15pm — to stand in for the “end of day” signal that you receive from seeing colleagues get up and leave the office…<br><em>— </em><a href="https://www.linkedin.com/in/kerry-sheldon/"><em>Kerry Sheldon</em></a></blockquote><blockquote>I still stick to a pretty strict schedule as if I were working in an office. 8ish-4:30ish<br><em> — </em><a href="https://www.linkedin.com/in/andylushman/"><em>Andy Lushman</em></a></blockquote><blockquote>Keeping a routine! Especially for later in the day to end the work day and be “home”<br><em> — </em><a href="https://www.linkedin.com/in/erinkrhodes/"><em>Erin Rhodes</em></a></blockquote><blockquote>Keep your “old” routine as much as possible. Avoid the temptation to sleep in until the second you have your first meeting and use the extra morning time to get organized for the day<br><em> — </em><a href="https://www.linkedin.com/in/kdubuisson/"><em>Kate Rose</em></a></blockquote><blockquote>Set working hours to define everyone’s availability. This addresses the variety of scheduling needs, work and life boundaries and the unexpected that can come up at any time but particularly with the backdrop of pandemic. As boring and adulting as it can be to put a structure around your time, it creates more flexibility.<br><em> — </em><a href="https://www.linkedin.com/in/sean-okeefe/"><em>Sean O’Keefe</em></a></blockquote><blockquote>Use your calendar to block off time. Don’t attend meetings that don’t have an agenda or obvious goal.<br><em> — </em><a href="https://www.linkedin.com/in/josh-mattila-0828b05/"><em>Josh Mattila</em></a></blockquote><h3><strong>Communication</strong></h3><blockquote>Our team set up a private channel with just our immediate teammates. Even when we’re pinging just one other teammate, we sometimes do it in that channel to mimic the interactions that we used to have in the office when turning to a teammate to ask a question. This gives other members the opportunity to “overhear” a conversation so they can chime in where they see fit, or simply stay in-the-know on random conversations happening within the team. Having this private channel in addition to our public team channel also makes it easier to freely share thoughts casually without worrying about spamming the wider audience of the public team channel.<br><em> — </em><a href="https://www.linkedin.com/in/allisonleong/"><em>Alli Leong</em></a></blockquote><blockquote>Our team has adapted to using the “/zoom” command [in Slack] at a moment’s notice to jump on a quick call to sort something out. Every once in a while we also hang out in a Zoom room while working on our own stuff, just to replicate the feel of being in the office together.<br><em> — </em><a href="https://www.linkedin.com/in/aprildagonese/"><em>April Dagonese</em></a></blockquote><blockquote>Standups are useful for starting the coordination of work every day. Consider Stand-down meetings at the end of the day if the team can’t have a consistent start time.<br><em>— </em><a href="https://www.linkedin.com/in/josh-mattila-0828b05/"><em>Josh Mattila</em></a></blockquote><blockquote>Every sprint, our squad has a standing meeting to review what technical decisions we need to make to address our current work. At this meeting, we also review what may be coming in the next few sprints, or larger architectural changes. It becomes a catch-all meeting for issues that crop up that don’t quite fit in with what we are currently doing, but that will help us in the future if we give them time and attention. Setting aside that space allows us to keep focus on current priorities. We are able to see both forests and trees and give the right level of attention when needed.<br><em>— </em><a href="https://www.linkedin.com/in/sean-okeefe/"><em>Sean O’Keefe</em></a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=57d87f556a54" width="1" height="1" alt=""><hr><p><a href="https://medium.com/extra-credit-by-guild/creating-a-working-from-home-environment-57d87f556a54">Creating a Working from Home Environment</a> was originally published in <a href="https://medium.com/extra-credit-by-guild">Extra Credit-A Tech Blog by Guild</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>