<?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[DataCamp - Medium]]></title>
        <description><![CDATA[Online learning platform with Python, R and SQL courses that focuses on building the best learning experience for Data Science - Medium]]></description>
        <link>https://medium.com/datacamp?source=rss----428316015075---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>DataCamp - Medium</title>
            <link>https://medium.com/datacamp?source=rss----428316015075---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 17:43:06 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/datacamp" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Why we’re building a collaboration product for data professionals and data fluent teams]]></title>
            <link>https://medium.com/datacamp/why-were-building-a-collaboration-product-for-data-professionals-and-data-fluent-teams-d981986f37e8?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/d981986f37e8</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[datacamp]]></category>
            <category><![CDATA[collaboration]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Jonathan Cornelissen]]></dc:creator>
            <pubDate>Tue, 16 Feb 2021 11:43:33 GMT</pubDate>
            <atom:updated>2021-02-16T11:40:57.806Z</atom:updated>
            <content:encoded><![CDATA[<h4>Doing data science work should be easy, enjoyable, and collaborative. That’s why we’re building a <a href="https://www.datacamp.com/workspace">collaboration product</a> for data professionals and data-fluent teams.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C9FZgTxSfI_LiynWNUdieQ.png" /></figure><p>When we started DataCamp, we wanted to make it easy, enjoyable, and affordable for everyone to learn data science online. Today, we’ve reached over 7 million learners and actively helped increase the data fluency of more than 1,600 organizations. Over 150,000 subscribers have access to DataCamp today and even more for free through <a href="https://www.datacamp.com/community/blog/datacamp-donates">DataCamp Donates</a> and <a href="https://www.datacamp.com/groups/education">DataCamp For The Classroom</a>. However, almost all of them face significant struggles as they start actually <em>doing</em> data work.</p><p><strong>The struggle of data professionals and data teams — and why it matters</strong></p><p>Just like <em>learning</em> data science, <em>doing</em> data science work should be easy, enjoyable, and collaborative. Even as the number of data professionals has exploded, there are almost no tools that are easy to use, easy to scale, and that take into account the reality that different people prefer different technologies and integrated development environments, or IDEs.</p><p>As a consequence, data scientists today are not as productive as they could be. Let’s face it, many of us data scientists waste a lot of time doing infrastructure and engineering work. Whether that’s setting up a machine, finding and accessing the right data, or deploying dashboards and reports. Not only do we hear these struggles from our learners and our customers, we’ve faced many of these struggles ourselves at DataCamp.</p><p>Moreover, data scientists don’t work alone. Doing data science work in a team setting should not be as painful as it is today. Most tools used by data scientists lack even the most basic collaboration features. Shockingly, collaboration among data teams often happens by sharing code and data through email, Slack, or other messaging tools. Version control is done through tools that are tailored to the workflow of software engineers and are not intuitive for data scientists — and don’t get me started on the lack of quality control on most data teams. Knowing which analysis, dashboard, or report is still accurate is critical for a team. The struggles of data professionals are real and typically start on day one of joining a team.</p><p>Organizations also lose as they lack a central place where all insights are gathered and where data professionals can learn from each other. Some insights live in notebooks, some in PDF reports, yet others in proprietary analytics tools like Looker and Metabase, and the list goes on. This leads to insights getting lost and efforts being duplicated, and creates frustration with business users and leadership interacting with data professionals.</p><p>This lack of proper tooling stands in the way of truly productive data professionals and data fluent organizations. While it is true that there’s been a lot of innovation for more BI-style users, this is often not true for data professionals who want to dig deeper and use the power of Python and R to generate their insights.</p><p><strong>What DataCamp is working toward</strong></p><p>We want to enable anyone, from anywhere and on any device, to access relevant data online, and to create visualizations, reports, and dashboards from their favorite tool or IDE and share it with fellow data professionals. We want to enable teams to seamlessly collaborate online in real time as well as asynchronously. We want teams to have basic version and quality control automatically built in. The end result should be a centralized insights layer for organizations of any size.</p><p>In other words, we want to make it as easy to do data science on your own (or your company’s) data as it is to take a DataCamp course.</p><p><strong>How we’ll get there</strong></p><p>Rome wasn’t built in a day. We’re currently building this product and are planning a phased rollout in the next 12 to 18 months.</p><p>First, we want to make it super easy for our learners to take their learnings from DataCamp — be it in the form of projects or courses — to a popular IDE like Jupyter. Working with the tools that are already widely used in industry and academia is a key step toward applying acquired skills in a realistic environment. In addition, users will be able to publish and share the work they’re doing in a workspace.</p><p>Next, we’ll make it seamless to connect your data and upload datasets, enabling users to start using DataCamp for day-to-day data science work. Furthermore, when users invite collaborators to review or edit their work, code and environment will be linked and self-contained, so the concerns around using the right package versions across projects and between different users becomes a thing of the past.</p><p>We’ll also focus on vastly improving real-time and async collaboration for data teams. Almost any team product today allows for real-time collaboration, commenting, and tagging. There’s no reason data teams should be deprived of that functionality, but they often are today.</p><p>Finally, as the product matures, we’ll continue to add features for organizations, leveraging the extensive user management architecture that we’ve already built for our enterprise learning platform. As an example, admins will only have to specify a database connection once and manage who has access to which data. On the receiving end, data scientists will have the right data connections in place when they start their analyses.</p><p>Improving data fluency and democratizing data science is the core of DataCamp’s mission. <a href="https://www.datacamp.com/workspace">This product</a> is a key step toward achieving that goal, and why we also believe the basic version of this product should be free forever.</p><p><em>If you’re interested in changing the way data teams collaborate, email us at </em><a href="/service/mailto:rocketship@datacamp.com"><em>rocketship@datacamp.com</em></a><em>. You can also reach out directly to </em><a href="https://www.linkedin.com/in/filip-schouwenaars-b576b74a/"><em>Filip Schouwenaars</em></a><em>, who’s taking the lead on building this new product, or our </em><a href="https://www.linkedin.com/in/dieterdemesmaeker/"><em>CTO, Dieter De Mesmaeker</em></a><em>. We’re always looking for talented engineers to join our team. </em><a href="https://boards.greenhouse.io/datacamp/jobs/2582866"><em>Apply now</em></a><em>!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d981986f37e8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/why-were-building-a-collaboration-product-for-data-professionals-and-data-fluent-teams-d981986f37e8">Why we’re building a collaboration product for data professionals and data fluent teams</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[2018 in Review]]></title>
            <link>https://medium.com/datacamp/2018-in-review-9ea13961999a?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/9ea13961999a</guid>
            <category><![CDATA[company]]></category>
            <dc:creator><![CDATA[DataCamp Team]]></dc:creator>
            <pubDate>Sat, 12 Jan 2019 14:43:58 GMT</pubDate>
            <atom:updated>2019-01-14T13:25:20.728Z</atom:updated>
            <content:encoded><![CDATA[<p>Happy 2019 from the DataCamp team!</p><p>At the start of this year, we’d like to take some time to look back on 2018. In this post, we’ll cover progress we’ve made helping to increase data fluency, building out our content library, improving the platform, and growing the DataCamp team.</p><h3>Spreading data fluency</h3><p>DataCamp is building the most comprehensive online learning resource for data science and analytics. We define success as helping individuals and companies to become more data fluent.</p><p>In 2018, DataCamp users spent an impressive<strong> 7.5 million hours actively learning</strong> on the platform. In that time, our users <strong>completed over 1.1 million courses</strong>, about 100 million exercises and over 65,000 projects to gain hands-on experience solving real-world problems.</p><p>We grew to over 85,000 individual subscribers in 2018, growing our total student base to 3.3 million, representing about 80% YoY growth. And our <a href="https://www.datacamp.com/groups/business">business offering</a> is now used by 1,200 companies around the globe, which represents significant growth compared to last year and making it the fastest growing segment of our revenue.</p><p>As part of our quest to improve data fluency around the world, we offer free <a href="https://www.datacamp.com/groups/education">DataCamp scholarships</a> to teachers and professors who want to use DataCamp in the classroom. We also provided scholarships to organizations that work with underrepresented and underprivileged groups in Africa and the Americas (reach out to <a href="/service/mailto:support@datacamp.com">support@datacamp.com</a> to learn more). This enabled us to provide <strong>DataCamp scholarships</strong> <strong>to over 100,000 people in 2018</strong>,<strong> </strong>including students learning in classes taught by<strong> more than 2,000 teachers and professors</strong>. This is more than all previous years combined, and we’d like to do even more in 2019.</p><p>In 2018, DataCamp also raised a $25 million financing round from <a href="https://www.spectrumequity.com/">Spectrum Equity</a> to continue our growth. You can read more about <a href="https://www.spectrumequity-blog.com/2018/12/19/datacamp-on-a-mission-to-increase-data-fluency-around-the-world/">Spectrum’s thesis for investing in DataCamp here</a>.</p><h3>Building out the content library</h3><p>In 2018 we doubled our content library to over <a href="https://www.datacamp.com/courses">200 courses</a>, <a href="https://www.datacamp.com/projects">50 projects</a> and <a href="https://www.datacamp.com/practice">30 practice modules</a>. All of this was possible because of the fine work of <a href="https://www.datacamp.com/instructors">more than 160 instructors</a>. Our library now includes all the major technologies (including <a href="https://www.datacamp.com/courses/tech:python">Python</a>, <a href="https://www.datacamp.com/courses/tech:r">R</a>, <a href="https://www.datacamp.com/courses/tech:sql">SQL</a>, <a href="https://www.datacamp.com/courses/q:spark">Spark</a> and <a href="https://www.datacamp.com/courses/tech:spreadsheets">Spreadsheets</a> — new!) and topics (like <a href="https://www.datacamp.com/courses/topic:machine_learning">Machine Learning</a> and <a href="https://www.datacamp.com/courses/topic:probablity_and_statistics">Probability and Statistics</a>) that individuals and teams interested in data science and analytics need to know.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Nn3zsbbia-20N15W" /></figure><h4>Instructor Community</h4><p>DataCamp owes much of its success to our amazing community of instructors. Without them, we wouldn’t have been able to double our course and project libraries in 2018! In the past year, we expanded our instructor pool to over 320 data professionals — including many who will release their first course in the coming months — and 50 who now have more than one course or project on our platform.</p><p>Our instructors are data experts that come from a variety of backgrounds, both academic and industry, including positions at Amazon, American Express, Deloitte Consulting, Nike, Spotify, Harvard, New York University, Northwestern, the University of Neuchatel and many more!</p><p>Diversity and inclusion remain important priorities for us as we broaden our curriculum. We now have more than 100 women and non-binary instructors, representing an increase from 19% in 2017 to 34% of our instructors today. In addition, we’ve expanded the range of nationalities represented in our instructor pool to 20 and we hope to continue expanding our global reach in the coming year.</p><p>As we head into 2019, advancing and strengthening our instructor community will continue to be very important for DataCamp. Look out for DataCamp networking events, workshops and more in 2019 across the globe. And if you’re interested in joining the ranks of our amazing instructors, please <a href="https://www.datacamp.com/create">apply</a>.</p><h4>Our Content Vision For 2019</h4><p>By the end of 2019, we want to grow our content library by more than 75%. You can expect more learning materials for Python, R, SQL and spreadsheets, as well as content in new curriculum areas such as data engineering, data science for executives, deep learning and AI, and even business intelligence tools. Additional resources will go into improving the tailored feedback messages you receive while taking DataCamp courses, projects and practice modes. Our ambition is to make your experience even more tailored to your learning needs.</p><h3>Improving the platform</h3><p>If you sign in to DataCamp today, the focus is no longer only on courses but on the full <a href="https://www.datacamp.com/community/blog/learn-data-science-practice-apply">learn-practice-apply philosophy</a>. We recommend what <a href="https://datacamp.com/projects">projects</a> match your current learning status best, what practice modes you should take to maintain data science fluency and what alternative courses you might like as well. In addition, we did a full upgrade of our search functionality, allowing you to better discover content in our library of more than 1,000 hours of data science learning content.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*wX3J_EVvOuU4-jJ-" /></figure><p><a href="https://www.datacamp.com/community/blog/practice-mode-arrives-on-mobile">We also extended our mobile application with practice mode</a> so that you can continue sharpening your data science skills no matter where you are. Now you can make your breaks and commutes even more productive by solving practice exercises for the courses you’ve taken on desktop or mobile. If you haven’t downloaded our mobile app yet, download it on <a href="https://itunes.apple.com/us/app/datacamp-learn-r-python/id1263413087">iOS</a> or <a href="https://play.google.com/store/apps/details?id=com.datacamp">Android</a> now.</p><p>Our business customers also got a <a href="https://www.datacamp.com/community/blog/datacamp-business-experience">whole new experience</a> with the launch of our revamped DataCamp for Business product, ideal for building data science and analytics skills within organizations of all sizes. It offers the ability to track team progress in your dedicated portal, set assignments, tailor development programs and set up SSO integrations. <a href="https://www.datacamp.com/groups/business">Learn more here</a>.</p><h3>Growing the DataCamp team</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/0*Jpe5fW5wL0Roakeb" /></figure><p>In 2018, we welcomed 65 new DataCamp team members as the team grew from 49 to 99, including 15 people who continued their careers elsewhere. It was exciting to welcome so many new and talented team members. At the same time, this led us to take a harder look at the culture we are building as well as ways to strengthen our internal policies and processes. We hired a Chief People Officer, invested in multiple diversity and inclusion trainings for all employees, strengthened the policies that govern how we work together, and conducted two anonymous diversity and inclusion surveys to help guide our continued growth in these areas. We know that we have more work to do in diversity and inclusion, and recognize that they are foundational elements for our continued growth and prosperity.</p><p>At the start of 2018, our team was very imbalanced in a few areas, including gender (predominantly male) and age (predominantly under age 35). Throughout the year, we worked toward creating a healthier balance in multiple areas with a specific focus on these two dimensions.</p><p>We’re proud that of the 30 people to join our team most recently (representing one third of the company), 50% identified as a woman or non-binary. For the team as a whole, we’re now close to 30% women and non-binary employees.</p><p>As it relates to increasing the experience level at DataCamp, we added multiple very experienced people at the leadership level in areas like HR, Sales, Marketing, Engineering, Product and Content. We feel confident that with this new layer of leadership, DataCamp is in a good position to continue to grow the team.</p><p>In 2018, we started running anonymous surveys with the goal to (1) measure self-reported demographics, (2) measure employee engagement and various aspects of inclusion, (3) determine where and how to improve our culture, and (4) benchmark our progress compared to previous results for DataCamp and other companies.</p><p>As data-driven decision making is a core value at DataCamp, we’ve decided to share the results from our latest diversity and inclusion survey. You can find a summary of the results below and an overview of the <a href="https://www.dropbox.com/s/gjv8n2ulvehc8my/Diversity_and_Inclusion_Survey_Dec_2018_All_Results_questions.pdf?dl=0">questions and responses here</a>. The survey summaries and the images below are benchmarked against <a href="https://academy.cultureamp.com/hc/en-us/articles/360000339229-2018-Diversity-and-Inclusion-Survey-Benchmark">CultureAmp’s 2018 Diversity And Inclusion Survey Benchmark</a>. More information on <a href="https://academy.cultureamp.com/hc/en-us/articles/360001319949-The-Methodology-Behind-the-Diversity-and-Inclusion-Survey">the methodology behind the survey can be found here</a>.</p><p>Figure 1 summarizes some of the key findings of the anonymous survey we ran in December of 2018. As you can see, overall employee engagement at Datacamp is 10 percentage points above the benchmark, which puts DataCamp in the 90th percentile of all companies that ran this survey. DataCamp is a leader in six areas, but we are still slightly below average on diversity (more on that later).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*SsAOhlzf9GuoJgDh" /><figcaption>Figure 1</figcaption></figure><p>Interestingly, CultureAmp also allows us to investigate what specific questions are key drivers of employee engagement or disengagement, which is shown in Figure 2.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5JeX7ImkW9ImtCsd" /><figcaption>Figure 2</figcaption></figure><p>DataCamp has been a mission-driven company from day one. We aim to hire people who agree that improving data fluency in the world at an affordable price is an important mission. Excitingly, Figure 3 highlights that almost all of our employees believe in this mission.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*HnU6uKdtzlzYPsxh" /><figcaption>Figure 3</figcaption></figure><p>While the average results are relatively positive, there are clear areas for improvement for the team overall and clear areas for improvement for certain segments of the team.</p><p>The lowest-rated statement in the survey is “DataCamp allows me to balance my work and personal life,” with a rating of 18 percentage points below the benchmark. Managing work-life balance during a period of accelerated growth has been a challenge for many team members. As a company, we believe this journey should be a marathon, not a sprint. We’ll be working across teams to create a healthier balance in 2019 (e.g. by holding managers accountable in this area, setting more realistic targets, etc.).</p><h4>Growing the team in 2019</h4><p>We’re planning to grow the team to about 175 this year, and are committed to continuing to invest to improve diversity and inclusion in 2019. Some of the key initiatives are below:</p><ul><li>Increase the percentage of women and non-binary employees to at least 35%, with a specific focus on increasing diversity at the leadership level.</li><li>Maintain employee engagement at the current level on average, while improving engagement for more junior employees and underrepresented groups by at least five percentage points. In key areas that contribute to overall engagement, we recognize that there are a range of experiences among employees, and we will work to improve the full employee experience across teams.</li><li>Continue to focus on improving other areas of diversity, such as race and ethnicity.</li><li>Continue to run anonymous diversity and inclusion surveys two times per year.</li><li>Continue to run multiple training programs, including unconscious bias training, respect in the workplace, cultural awareness and training on people management.</li><li>Continue to invest in top-of-funnel initiatives that increase diversity (e.g. by posting on specific job boards and attending events focused on underrepresented groups within the tech community).</li><li>Work as a team to document our culture in more detail and share that document with the community.</li><li>Continue to provide support for up to 5% of employees to participate in the DataCamp Exchange program every year. This program offers the opportunity to employees to move between our office locations (New York, London and Belgium) and stay for one year. This enables employees to be exposed to different cultures and helps DataCamp to build bridges between its offices.</li><li>Continue to improve cultural diversity by investing in visa sponsorships in the different countries in which we work. We currently have a team comprised of 15 different national origins.</li></ul><p>We’re very excited about 2019 here at DataCamp. If you’d like to join our team, please look at our <a href="https://www.datacamp.com/jobs">open positions</a>.</p><p><em>The DataCamp Team</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9ea13961999a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/2018-in-review-9ea13961999a">2018 in Review</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Making Sense of Design Systems]]></title>
            <link>https://medium.com/datacamp/making-sense-of-design-systems-ffd38c6151d3?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/ffd38c6151d3</guid>
            <category><![CDATA[datacamp]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[company]]></category>
            <dc:creator><![CDATA[Kristof Houben]]></dc:creator>
            <pubDate>Mon, 23 Apr 2018 10:54:01 GMT</pubDate>
            <atom:updated>2018-04-23T10:55:36.737Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3h8QfJCt5UZU_XvToEA1ug.gif" /></figure><p>At <a href="https://www.datacamp.com/home">DataCamp</a> our <a href="https://www.datacamp.com/community/blog/datacamp-the-plan">mission</a> is to <strong>democratize data science education for everyone</strong>, and with over 2 million students worldwide we’re making rapid progress. Operating at this growth rate requires a lot of effort from everyone within the company and introduces a lot of exciting new challenges.</p><p>For the design team it means that we’re shifting from page-by-page design to a more scalable way of designing. Reusing existing components, reducing development time, improving usability, and increasing accessibility are becoming more and more important.</p><p>This article covers how we are starting to make sense of a design system.</p><h3>What’s a design system?</h3><p>A <strong>design system</strong> brings together all the components that are needed to <strong>distribute</strong> and <strong>reproduce</strong> a design solution. It’s not something new or exclusive to digital product design — a lot of examples can be found throughout history.</p><p>For example, the first Chinese empire had the problem that archers couldn’t use arrows from their comrades. Each arrow was specifically designed for a certain bow, so other arrows simply wouldn’t fit. The first Chinese emperor, <a href="https://www.brainpickings.org/2013/03/15/alice-rawsthorn-hello-world/">Ying Zheng</a>, <strong>standardized the way arrows were designed</strong> and it had a huge impact on their military conquests.</p><p>Back in the 50’s, each station of the NYC Transit System had its own signage. It caused a lot of confusion because every single station had their own way of navigating. As a result, a lot of people were hesitant to use the transit system because they <strong>didn’t understand how it worked</strong>. <a href="https://en.wikipedia.org/wiki/Massimo_Vignelli">Massimo Vignelli</a> and <a href="https://en.wikipedia.org/wiki/Bob_Noorda">Bob Noorda</a> created the <a href="https://standardsmanual.com/products/nyctacompactedition">NYC Transit Authority — Graphics Standards Manual</a>, which documented how signs should be made and how they are supposed to be placed within stations. Over time this system has remained more or less the same and <strong>it’s still actively being used today</strong>.</p><h3>Why should we have a design system?</h3><p>When every page is designed in isolation, a lot of components have duplicates or similar alternatives. It makes a digital product hard to understand by users and hard to maintain by engineering. At DataCamp, we discovered over 20 different buttons throughout our product, each with their own quirks (different hover states, inconsistent disabled states, etc). It’s hard to keep track of all these components when a product is growing rapidly and it becomes near impossible to maintain. That’s where a design system kicks in.</p><p>For a digital product, a design system describes the <strong>user interface components</strong> that can be used to <strong>create an interface</strong> that solves a problem and will result in a <strong>cohesive experience</strong> for the user.</p><p>When a design system is executed well it will result in a cascading effect that …</p><ul><li>Decreases miscommunication</li><li>Decreases implementation time</li><li>Decreases maintenance work</li><li>Decreases bugs</li><li>Decreases support tickets</li><li>Increases quality</li><li>Increases efficiency</li><li>Increases longevity</li><li>Increases learnability</li><li>Increases usability</li><li>Increases accessibility</li><li>Increases innovation</li><li>Increases focus</li><li>Increases sales</li></ul><h3>It’s not easy …</h3><p>Creating, maintaining and enforcing a design system requires a lot of <strong>discipline</strong> and a <strong>continuous effort</strong> for everyone that’s part of the creation process. It’s not just design; engineering and product management are also important stakeholders. They all should be able to ask tough questions like: “Why are we building this?”, “Can we reuse this component somewhere else?”, “Do we already have something similar?”, “Does this custom component add enough benefit opposed to the effort we put in?”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*wz94rdxK_kfsyXt9.png" /></figure><h3>Design System 101 at DataCamp</h3><h4>1. Create A Component Inventory</h4><p><strong><em>Goal</em></strong><em>: Make sure everyone knows the existence and status of a component.</em></p><p>When I started at DataCamp there was a <a href="https://sketchapp.com/">Sketch</a> file that contained colors, icons, fonts and some basic components. It was far from complete but it was a good starting point. Front-end engineering also had a collection of components they used across projects.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*uQY76NTPKuL3p5LG.png" /></figure><p>Together we <strong>created an inventory</strong> in <a href="https://airtable.com/">Airtable</a> based on the components that are used in production. This gave the design team an idea of what already existed and prevented us from unknowingly creating new components. It also gave us an insight on how many similar components we had and how bad the situation really was.</p><h4>2. Use Naming Conventions</h4><p><strong><em>Goal</em></strong><em>: Avoid miscommunication between Design and Engineering.</em></p><p>When we started comparing components we noticed Design and Engineering were not speaking the same language. For example the ‘primary color’ that was defined by design was the ‘secondary color’ in code. Things like this can lead to miscommunication when teams start to scale and causes confusion and frustration. We made sure that the same lingo is being used across the different teams.</p><h4>3. Optimizing The Design Workflow</h4><p><strong><em>Goal</em></strong><em>: Avoid losing work and have a transparent workflow.</em></p><p>Working together as a design team introduces its own set of challenges. We used to store designs in Dropbox but when our team started to scale, the risks of losing work and the effort to keep track of changes increased dramatically.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*v4i7-kUKGVdQzuvF.png" /></figure><p>Our design team was really open for a workflow change and we started using <a href="https://www.goabstract.com/">Abstract</a>. It enabled us to properly version our designs, have a transparent workflow, and have more confidence while designing.</p><h4>4. Better Documentation + Onboarding</h4><p><strong><em>Goal</em></strong><em>: Faster employee onboarding and prevent communication overhead.</em></p><p>We started documenting our process and design system using <a href="https://slite.com/">Slite</a>. If we have to explain something more than once it should be <strong>documented and shared</strong> with the team. This avoids a lot of communication overhead and prevents information from getting lost.</p><h4>5. Design Reviews</h4><p><strong><em>Goal</em></strong><em>: Setting expectations between Design and Engineering.</em></p><p>It’s important that design reviews happen in <strong>both directions</strong>. Designers should be able to review front-end, but front-end should also be able to review design. Designers have this urge to create something cool and new. Engineers should be able to ask tough questions like: “Why are we building something custom?”, “Can’t we reuse this component?”, “Can we make this component better?”, “Is this component worth the extra development effort?”. It’s about <strong>justifying decisions</strong> and <strong>setting expectations</strong> to everyone involved.</p><h3>Conclusion</h3><p>We just started putting together our design system and we’re still trying to make sense of all the duplicate components we found throughout our product. There’s still a long road ahead but with every component we touch we’re creating a better, more consistent, and easier to learn product for our customers.</p><p>We’re curious how other organizations make sense of a design system. Send me an <a href="/service/mailto:kristof@datacamp.com">email</a> if you’d like to share your experience.</p><p><em>At DataCamp we’re always looking for talented people to join our team. Interested? </em><a href="https://www.datacamp.com/jobs"><em>Check our open positions</em></a><em>.</em></p><p><em>Originally published at </em><a href="https://www.datacamp.com/community/blog/design-systems-datacamp"><em>www.datacamp.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ffd38c6151d3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/making-sense-of-design-systems-ffd38c6151d3">Making Sense of Design Systems</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Duolingo-style Learning for Data Science: DataCamp for Mobile]]></title>
            <link>https://medium.com/datacamp/duolingo-style-learning-for-data-science-datacamp-for-mobile-3861d1bc02df?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/3861d1bc02df</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[rstats]]></category>
            <category><![CDATA[mobile]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Boris Gordts]]></dc:creator>
            <pubDate>Tue, 23 Jan 2018 13:26:28 GMT</pubDate>
            <atom:updated>2018-01-23T13:26:27.940Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Data science is the future, but learning it takes a lot of time. Time you don’t have? Wrong. With </em><a href="https://www.datacamp.com/mobile"><em>DataCamp for Mobile</em></a><em>, you master data science while on the train or on the bus (or on the toilet, as we’ll explain). You don’t have to be a math genius either. DataCamp’s Learning by Doing methodology makes learning it intuitive for everyone.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*jg2MrSpS5tFHHMwM.png" /><figcaption>DataCamp for Mobile</figcaption></figure><h3>Introducing DataCamp for Mobile — or how to learn data science from the toilet</h3><p>At DataCamp, we’re dedicated to bringing you the best possible learning experience for data science. At the core, there are our 100+ intuitive web courses teaching the fundamentals of R, Python, and SQL as well as more advanced topics like Machine Learning and Deep Learning. Furthermore, to attain true proficiency, our Practice Mode lets you train what you’ve learned before, and with Projects you get hands-on experience solving real-world problems.</p><p>However, a big problem for many of our users is finding the time to learn and practice. That’s why we wanted to make a version of DataCamp that better fits a busy schedule. But how? The answer is simple: better use the time that is otherwise lost. Did you know you spend 20 hours per year on the toilet? That’s more time than you need to complete our <a href="https://www.datacamp.com/tracks/python-programming">Python Programming skill track</a>!</p><blockquote><strong><em>Did you know you spend 20 hours per year on the toilet? That’s more time than you need to complete our Python Programming skill track!</em></strong></blockquote><p>DataCamp for Mobile (available in the <a href="https://itunes.apple.com/us/app/datacamp-learn-r-python/id1263413087?mt=8">Apple App Store</a> and <a href="https://play.google.com/store/apps/details?id=com.datacamp&amp;hl=en">Google Play Store</a>) offers courses on Python, R and SQL that help you master the practical skill set to do data science. Taught in a Duolingo-style way, you’ll still learn the same concepts as you would in our web courses. It also contains a practice mode to train what you learned before.</p><p>Courses are divided into 5 minute lessons that teach you one concept at a time. Each lesson gently introduces data science concepts without making you read tons of text. Instead, all material is presented as an exercise. Starting from zero, each exercise builds on the next to break down even the most complex data science concepts in understandable pieces. This learning by doing methodology is the core of the DataCamp experience and the foundation of its success.</p><blockquote><em>Starting from zero, each exercise builds on the next to break down even the most complex data science concepts in understandable pieces. This learning by doing methodology is the core of the DataCamp experience and the foundation of its success.</em></blockquote><p>Our users seem to like it so far: the app was downloaded almost 100.000 times in only a few weeks, and the average app store rating is 4.5/5.</p><h3>The Learning Experience</h3><p>There are two main challenges when teaching on mobile: the small screen size and the limited attention span of mobile users compared to web users. That’s why there is only one way to make the mobile app successful: creating a learning experience that is as engaging as a game.</p><p>Every 5-minute lesson is divided into a set of exercises and is designed to only take a couple of minutes. Here are a couple of examples of the exercises you can expect:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*Q5Dq-p2R90-qvzKp4ap_IA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*P7Zaa-ro1bAf2U-0IZj--Q.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*9H3fs_WpUbRBX1hfwGqTUg.png" /><figcaption>Some examples of the DataCamp for Mobile learning interface.</figcaption></figure><p>To make the learning experience as engaging as possible, we want to bring you into the “learning flow”. Whenever you start a lesson, every element on the screen is focused on either teaching or motivating you about the progress you’ve made.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*yA0IZ8DH0u9ZkDhBT1l22Q.gif" /><figcaption>An example of the learning flow</figcaption></figure><p>As you can see, we paid a lot of attention to the details to build up this flow:</p><ul><li>All elements on the screen have a fixed location: first a little explanatory text, then the question, then the answer options and finally the check button.</li><li>The text has the size of a tweet so you never have to read too much.</li><li>Every exercise has the same sequence of actions: read the question, solve it, hit check, absorb feedback, hit continue, repeat. By keeping this flow consistent, you get in a rhythm that brings you “in the zone”.</li></ul><p>Of course, it is inevitable to make a mistake from time to time. That’s why we add a personalized feedback message to explain what is wrong. Whenever you answered a question incorrectly, it will come back later in the lesson.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*bHcgeCI8tS2rmOo0gWGnoQ.png" /><figcaption>When you get an exercise wrong, the app will tell you why.</figcaption></figure><p>DataCamp for Mobile also offers a practice mode, which allows you to train what you’ve learned before. To help you get a habit of practice, we are gamifying many aspects of the app. For example, we added a daily streak indicator that shows how many days in a row you learned data science. Expect more gamification features in the near future!</p><h3>What’s next?</h3><p>Only being a couple of months old, there are still many ways to improve DataCamp for Mobile.</p><ul><li><strong>Add more courses</strong>. Recently, we added Intermediate Python and Introduction to SQL to the app, and many more courses are in the works. We’re also reaching out to external instructors, to bring you the highest quality courses by experts in the field (if you are interested to make a course or you know someone, please don’t hesitate to reach out to mobile@datacamp.com!).</li><li><strong>Bring all practice content we have on web to mobile</strong>. Practice mode already <a href="https://www.datacamp.com/community/blog/introducing-daily-practice-mode">exists on web for a while</a>, with exercises for 17 courses! Our goal is to bring all these practice exercises to mobile in the coming months.</li><li><strong>Improve the existing content and user experience</strong>. Being a data-driven company, we will do a thorough data analysis of our current content and make improvements where necessary. Also, a thorough redesign of the course navigation screens is on the roadmap. Finally, we’ll be adding — much-requested — sounds for good and bad answers.</li></ul><p>Would you like to see something else? Please don’t hesitate to share any feedback you have with us in the comments below or at mobile@datacamp.com!</p><h3>Conclusion</h3><p>DataCamp for Mobile brings DataCamp’s intuitive learning experience to become data fluent to mobile. This way, we hope to make data science education accessible to people with a busy schedule. Our users seem to like it as well! The app was downloaded almost a 100k times in just a few weeks and has an average app store rating is exceeding 4.5/5.</p><p>So, next time you are swiping through Facebook or Instagram (maybe during a restroom visit?), consider whether you don’t want to open DataCamp for Mobile instead ;) Happy Learning!</p><p><em>Are you interested in mobile or online education? Join our team! We are looking for developers and many other roles. Apply now at </em><a href="https://www.datacamp.com/careers"><em>https://www.datacamp.com/careers</em></a><em>.</em></p><p><em>Originally published at </em><a href="https://www.datacamp.com/community/blog/mobile-data-science"><em>www.datacamp.com</em></a><em>. Written by Jens Claes and Boris Gordts.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3861d1bc02df" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/duolingo-style-learning-for-data-science-datacamp-for-mobile-3861d1bc02df">Duolingo-style Learning for Data Science: DataCamp for Mobile</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DataCamp: The plan]]></title>
            <link>https://medium.com/datacamp/datacamp-the-plan-7ff04346fd6b?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/7ff04346fd6b</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[edtech]]></category>
            <category><![CDATA[company]]></category>
            <category><![CDATA[education]]></category>
            <dc:creator><![CDATA[Jonathan Cornelissen]]></dc:creator>
            <pubDate>Wed, 29 Nov 2017 18:55:12 GMT</pubDate>
            <atom:updated>2019-07-26T16:39:17.933Z</atom:updated>
            <content:encoded><![CDATA[<p>In a previous post, I gave some context on <a href="https://www.jonathancornelissen.com/post/why-we-started-datacamp"><em>why we started DataCamp</em></a> at a macro level. This post describes the key steps in building out DataCamp, i.e. the plan.</p><p>First of all, let’s fast forward to what we’re trying to achieve here in the long run: <em>“DataCamp’s mission is to help individuals and companies at every step of their journey to become data fluent by building the </em><strong><em>smartest</em></strong><em> </em><strong><em>data science education platform </em></strong><em>out there.” .</em></p><p>To be more specific, the end goal is to:</p><ul><li>Enable <strong>individuals</strong> to learn data science in a fun and engaging way, stay up to speed with the latest and greatest in data science land, get mentorship and help in an efficient way during their learning process, have their skill level certified, and help them advance in their career.</li><li>Empower <strong>companies</strong> to retrain their workforce effectively, understand the skills of their workforce, find and screen the right applicants for jobs.</li></ul><p>At a high level, there’s 5 steps to reach that end goal:<br><em>Step 1: Build the best learning platform for data science <br>Step 2: Attract and retain students<br>Step 3: Attract and retain the best instructors<br>Step 4: Help companies to retrain their workforce<br>Step 5: Expand our offering</em></p><p>In what follows, we take a deeper dive into the steps of the plan. All of these steps are required to reach the end goal. They are sequential in the sense that we want to get to a decent level for every step, before we move to the next step. Focus is crucial as explained in step 5. That said, it’s extremely important to understand that we need to keep executing on the first steps as we move to new ones.</p><p><strong><em>Step 1: Build an engaging and effective platform to learn data science </em><br></strong>We’ve built 3 different (sub)products that help students learn data science in an engaging and effective way:</p><ol><li><em>Interactive courses</em></li><li><em>Practice mode</em></li><li><em>Projects</em></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hGUao7UjIsjONCVlvZoKdA.png" /><figcaption>One interactive exercise interface — Students use R in the comfort of their browser and get personalized feedback while solving challenges.</figcaption></figure><p><strong><em>Interactive courses: </em></strong>Our focus on data science enables us to build interactive exercise interfaces that work really well for learning data science. Students solve challenges with R/Python/.. in the comfort of their browser. Furthermore, we’ve created custom technology to give personalized and instant feedback to students when they are solving challenges. We want our students to spend at least 70% of the time with their hands on the keyboard, actively solving challenges and receiving instant and personalized feedback. Not only does this lead to higher engagement (many of our courses have completion rates of over 50%), we believe it is a more effective way of learning for most students.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Iq1Q7Bn3mPnGT1_00pnF0w.png" /><figcaption>Practice mode interface — Students build fluency through simplified challenges</figcaption></figure><p><strong><em>Practice mode: </em></strong>Our interactive courses are great to teach students new skills. The courses are not very helpful to build fluency and retention of knowledge. In contrast, for practice mode, we encourage students to practice regularly as <a href="https://en.wikipedia.org/wiki/Spaced_repetition">spaced repetition</a> leads to better knowledge retention. Furthermore, the somewhat simplified bite-sized challenges help students to build fluency. Practice mode is a fairly new part of our product: the goal is to become really good at understanding someone’s skill level and offer the right amount of practice on the right topics at the right time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1008/1*obJYJznwwNgZYVhcYhopaA.png" /><figcaption>DataCamp on Mobile — have a look at the <a href="https://play.google.com/store/apps/details?id=com.datacamp&amp;hl=en">Android</a> or <a href="https://itunes.apple.com/us/app/datacamp-learn-r-python/id1263413087?mt=8">iPhone app</a>!</figcaption></figure><p>Furthermore, we’re investing heavily in a <strong>mobile app</strong> that allows our students to learn new topics and practice data fluency on the go.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6AupWRMFEV3o_0iEwuaJ1g.png" /><figcaption>Project interface — Student’s last step towards doing data science projects independently</figcaption></figure><p><strong><em>Projects: </em></strong>The previous 2 elements of our learning product help to: (i) gain understanding of a topic, (ii) become fluent in applying it and (iii) have a high probability of retaining that knowledge. There’s still something missing though. Our students want to be able to make better decisions based on data and gain insight from data. They want to tackle real data science projects. In the next few weeks, we’ll launch a new part of the product: <em>projects. </em>Projects help students transition into the real world of data science. The goal here is to bring the learning experience as close as possible to what they will be doing in actual data science projects.</p><p>By the end of 2017, we expect these 3 core components of the learning experience to reach a decent level of maturity from a product perspective. In addition to further improving them, we want to invest in:</p><ul><li><em>Gamification:</em> Learning should be fun. DataCamp’s current gamification system with experience points will get a complete revamp.</li><li><em>Mobile: </em>Many of our students are busy. A mobile learning experience allows them to learn on the go. We’d like to push the boundaries and we’re currently rethinking how learning for data science can happen on a mobile phone.</li><li><em>Resource features: </em>As our content library grows, we need better ways for students to navigate through the content library, and enable them to use the platform as a resource as well.</li></ul><p><strong><em>Step 2: Attract and retain students</em></strong></p><p>When we built the first version of the platform, we faced the chicken or egg problem: If you don’t have students, no one is interested in building courses, if you don’t have courses you certainly won’t attract any students.</p><p>We did two things to overcome this challenge. First, we created <a href="https://www.datacamp.com/courses/free-introduction-to-r">a course</a> for absolute beginners and made it available for free. As people started linking to that course, it started ranking in top position for various “learn R” related search terms. We intend to keep releasing free courses on topics that matter to beginners. Second, we built complements to existing courses on <a href="https://www.coursera.org/specializations/executive-data-science">Coursera</a> to improve their lab/exercise components, and set up a similar collaboration with Microsoft for their <a href="https://medium.com/u/6e8411791314">edX</a> courses (on <a href="https://www.edx.org/course/introduction-r-data-science-microsoft-dat204x-6">R</a>, <a href="https://www.edx.org/course/introduction-python-data-science-microsoft-dat208x-7">Python</a> and <a href="https://www.edx.org/course/introduction-python-data-science-microsoft-dat208x-7">SQL</a>). These collaborations helped to get the word out on DataCamp initially.</p><p>DataCamp now has over 1.7 million students. We’re ready to welcome millions more in 2018. A few of our current initiatives to welcome new students:</p><ul><li><a href="https://www.datacamp.com/groups/education">Teachers who want to use DataCamp in the classroom</a> can do so entirely for free. Over 1,000 teachers are currently using DataCamp in the classroom (and tens of thousand of students). In addition to access to the entire course library, they get access to numerous tools that help them use DataCamp in the classroom.</li><li>Complete beginners can enjoy our introductory courses on <a href="https://www.datacamp.com/courses/intro-to-python-for-data-science">Python</a>, <a href="https://www.datacamp.com/courses/free-introduction-to-r">R</a> &amp; <a href="https://www.datacamp.com/courses/intro-to-sql-for-data-science">SQL</a> for free (Shell &amp; Git are coming soon).</li><li>Everyone can enjoy the <a href="https://www.datacamp.com/community/tutorials">tutorials</a> and <a href="https://www.datacamp.com/community/blog">data science blog posts</a>.</li><li>Every first chapter of every course is available for free, so student can try before they buy a subscription.</li><li>We allow everyone to <a href="https://www.datacamp.com/teach/">create DataCamp courses</a>. Have a look at the list of <a href="https://www.datacamp.com/community/open-courses">these open courses</a>. Some of them are translations of existing DataCamp courses in <a href="https://www.datacamp.com/community/open-courses/introduction-a-r">French</a>, <a href="https://www.datacamp.com/community/open-courses/introduccion-a-r">Spanish</a>, <a href="https://www.datacamp.com/community/open-courses/r-%E8%AA%9E%E8%A8%80%E5%B0%8E%E8%AB%96#gs.Km2bhjQ">Chinese</a>, <a href="https://www.datacamp.com/community/open-courses/introduzione-a-r">Italian</a>, <a href="https://www.datacamp.com/community/open-courses/h%C6%B0%E1%BB%9Bng-d%E1%BA%ABn-c%C6%A1-b%E1%BA%A3n-v%E1%BB%81-r">Vietnamese</a>, <a href="https://www.datacamp.com/community/open-courses/einfuhrung-in-r">German</a> &amp; <a href="https://www.datacamp.com/community/open-courses/introducao-a-r#gs.VfwEqvY">Portugese</a>.</li><li>A mobile app with free introductory courses and practice sessions (download for <a href="http://play.google.com/store/apps/details?id=com.datacamp">Android</a> or <a href="https://itunes.apple.com/us/app/datacamp-learn-r-python/id1263413087?mt=8">iPhone</a>).</li></ul><p>We love offering free services to our students. That said, we also want to build a large and sustainable business. This is the best way to ensure that we can keep innovating and providing value to our students. Our business model is simple: to get access to the entire content library students pay an affordable fee of $29/month or $300/year. This is either on par or much lower than alternative options. We are committed to maintaining a low price point as we believe in the importance of <em>data fluency </em>for a large group in society. There’s room for improvement here though: ideally, we would have a price point adjusted to the GDP per capita in a country for example (we have subscribers from most countries in the world).</p><p>As we’re getting better at attracting students, our focus should be shifting more towards student retention. Supporting students in their journey certainly helps to improve retention. Unfortunately, due to our rapid growth, we’ve not always been able to provide excellent support. By the end of 2017, we’ll have a dedicated customer support and a customer success team. Not surprisingly, our data shows that product improvement and the addition of more content also has a positive effect on student retention.</p><p><strong>Step 3: Attract and retain the best instructors</strong></p><p>As mentioned above, we believe creating a network of <a href="https://www.datacamp.com/instructors">expert instructors</a> to teach our data science courses. Instructors build DataCamp courses for various reasons. The three main reasons being:</p><ol><li><em>Impact</em>. DataCamp has become the most effective way to reach over 1.7 million students eager to learn data science. DataCamp’s reach helps instructors build their personal brand as well.</li><li>Belief in the <em>educational vision</em>. Many educators care. They are looking for novel and innovative ways to teach data science topics. Our focus on data science enables us to offer the best learning tools.</li><li><em>Financial</em>. We share revenue with instructors in a unique way: <em>instructors are paid for every student that finishes a course. </em>This ensures that incentives are aligned around creating engaging courses that our students enjoy.</li></ol><blockquote>Compared to writing a book, our instructors can reach more students, teach more effectively and make more money.</blockquote><p>Making sure our instructors get value from creating content on DataCamp is essential for the long term success of the platform. Therefore, we have a content team focused on (i) finding the best people in the world to join that network, (ii) supporting instructors to teach according to best practices and (iii) providing practical support and insightful feedback to help them create and maintain the best data science courses out there.</p><p>As there’s a learning curve to creating DataCamp content, we want instructors to create multiple courses. The following external <a href="https://www.datacamp.com/instructors">instructors</a> already created multiple courses on the platform: <a href="https://www.datacamp.com/instructors/garrettgrolemund">Garrett Grolemund</a>, <a href="https://www.datacamp.com/instructors/continuum">Dhavide Aruliah</a>, <a href="https://www.datacamp.com/instructors/cwickham">Charlotte Wickham</a>, <a href="https://www.datacamp.com/instructors/rickscavetta">Rick Scavetta</a>, <a href="https://www.datacamp.com/instructors/bbaumer">Ben Baumer</a>, <a href="https://www.datacamp.com/instructors/bois">Justin Bois</a>, <a href="https://www.datacamp.com/instructors/drobinson">David Robinson</a>, <a href="https://www.datacamp.com/instructors/jasonamyers">Jason Myers</a>, <a href="https://www.datacamp.com/instructors/dtkaplan">Daniel Kaplan</a>.</p><p>In addition to forging a network of instructors, we’re designing a scalable process for the creation of high-quality content. Technology is key in this context. We have invested heavily in technology that makes it easy to create interactive courses on DataCamp. That said, there’s still a lot of room for improvement on <a href="https://www.datacamp.com/teach">DataCamp’s authoring environment</a>. It’s a core part of our strategy to make it very easy and scalable to create content for DataCamp’s platform (see <a href="https://medium.com/datacamp/what-world-war-2-can-teach-you-about-product-management-3053138f8be8">this post</a> as well). With this network of instructors and efficient authoring technology, we believe it’s realistic to have about 1,000 content modules (=courses/projects/practice modes) by 2019.</p><p>It doesn’t stop at authoring great content though. Students have completed over 70 million exercises on DataCamp. We can learn a lot from that data, and our content team and instructors can and do use that data to improve the content. We have an unfair advantage here as a company, since a large percentage of DataCamp employees and instructors actually know how to work with data well.</p><p><strong>Step 4: Help companies to retrain their workforce</strong></p><p>Most people learn data related skills in a professional context. Therefore, the end game is to offer scalable training solutions to companies. It helps increase DataCamp’s impact and provides a more stable revenue stream.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NOWXrQeZDGm07FwAWqutoQ.png" /><figcaption>DataCamp’s business product helps organizations become more data literate.</figcaption></figure><p>With already over 1.7 million registered students, we can rely on engaged professional students to help bring DataCamp into their organization. Often, we start by training smaller groups of people (20–50). Their enthusiasm then spreads within the organization and other groups sign up for a <a href="https://www.datacamp.com/groups/business">DataCamp for business license</a>. Ultimately, we want to help the entire organization become more data literate through enterprise licenses. If you’re curious how <a href="http://datacamp.com/">DataCamp</a> is helping companies, have a look at the <a href="https://www.datacamp.com/groups/business">case studies </a>on our business page.</p><p>As we move towards helping organizations retrain their workforce, we’re also expanding our product offering to cater to the needs of larger organizations. In the next months (years), we’ll continue to launch product features focused at helping organizations such as:</p><ul><li>Ability to create custom learning paths for groups and individuals</li><li>Detailed reporting on what and how employees learn, their skill level, what they struggle with, etc.</li><li>..</li></ul><p><strong>Step 5: Expand our offering</strong></p><p>Many students, employees &amp; investors have asked me some version of “Why is DataCamp <em>not</em> doing XYZ”? With XYZ being, anything from mentorship, to recruiting, to certification, etc. The real question is often not <em>why not but when is it appropriate.</em></p><p>In the first years of building DataCamp, we deliberately want to remain hyper focused on building a learning resource with the best instructors. We believe that creating a great learning resource for data science is not only what’s most missing in the market right now. It’s also the most defensible asset. The defensibility follows from: (i) the fact that it takes a lot of time, effort and the right team to create such a resource (content+product), and (ii) that it’s an asset that captures people and companies at the start of their journey towards data fluency.</p><p>When we feel we master the first four steps, we want to expand our offering (at the time of this writing, DataCamp is gradually shifting focus to <em>step 4</em>). The importance of certification for skill level was discussed in <a href="https://medium.com/@jonathan_43475/context-on-why-we-started-datacamp-d4df02fcc87c">a previous blog post</a>, and seems the most likely candidate for initial expansion of our offering. We know our individual subscribers and corporate clients want mentorship, help in their job search, etc., so we won’t stop there. <em>DataCamp wants to help as they progress the journey towards more data fluency.</em></p><blockquote>So.. In a nutshell <br>1. Build the best learning platform for data science <br>2. Attract and retain students<br>3. Attract and retain the best instructors<br>4. Help companies to retrain their workforce<br>5. Expand our offering</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7ff04346fd6b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/datacamp-the-plan-7ff04346fd6b">DataCamp: The plan</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Context on why we started DataCamp]]></title>
            <link>https://medium.com/datacamp/context-on-why-we-started-datacamp-d4df02fcc87c?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/d4df02fcc87c</guid>
            <category><![CDATA[education]]></category>
            <category><![CDATA[company]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[edtech]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Jonathan Cornelissen]]></dc:creator>
            <pubDate>Wed, 29 Nov 2017 18:54:25 GMT</pubDate>
            <atom:updated>2019-07-17T01:28:30.332Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://www.jonathancornelissen.com/post/why-we-started-datacamp">This story has been moved to a different location.</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d4df02fcc87c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/context-on-why-we-started-datacamp-d4df02fcc87c">Context on why we started DataCamp</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Technical Vision Blog — Part 3]]></title>
            <link>https://medium.com/datacamp/technical-vision-blog-part-3-cabde9c7a041?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/cabde9c7a041</guid>
            <category><![CDATA[docker]]></category>
            <category><![CDATA[scalability]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[company]]></category>
            <dc:creator><![CDATA[Dieter De Mesmaeker]]></dc:creator>
            <pubDate>Fri, 25 Aug 2017 14:35:26 GMT</pubDate>
            <atom:updated>2017-08-25T14:35:10.604Z</atom:updated>
            <content:encoded><![CDATA[<p>As our engineering team started growing, so did our number of microservices. Initially we started off with 3, now we have closer to 30. We <a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">believe in giving our engineers ownership</a> over their services. This means that they have a high level of freedom on how to structure their application, which programming languages to use, which API to expose, etc.</p><p>In this part we will discuss how we plan on creating more consistency between services, without limiting the ownership of our engineers</p><h3>Challenge 3: Scaling consistency between services</h3><p>We have grown rapidly in the last few years. As a result, we have taken some technical debt when wanting to launch new services ASAP. Some of our stack is running on <a href="https://www.heroku.com/">Heroku</a> (mostly lower traffic services), others on <a href="https://aws.amazon.com/ecs/">Amazon EC2 Container Service (ECS)</a> (direct or by using <a href="https://aws.amazon.com/elasticbeanstalk/">Amazon Elastic Beanstalk</a>). This seems fine, however when auditing the services we found some irregularities:</p><ul><li>Some services did not have autoscaling.</li><li>Other services relied on manual deployment to staging and dev.</li><li>Monitoring and alerting was configured for most of the critical services, yet this was not the case for all of them.</li><li>The flow for deploying to production and rolling back was very different depending on the service.</li><li>Some applications use Cloudwatch for logging, others rely on Papertrail, others have no searchable logs, etc.</li></ul><p>As a result some engineers need to adapt their applications to Heroku, others need to write Dockerfiles, some need to learn how to configure ECS, etc.</p><p>We have chosen to enforce consistency by moving all microservices to our own cluster:</p><ul><li>Applications need to run inside a <a href="http://docker.com/">Docker</a> container</li><li>Container orchestration handles by <a href="https://kubernetes.io/">Kubernetes</a></li><li>Automated deployment / rollback using <a href="https://jenkins.io/">Jenkins</a> (and <a href="https://jenkins.io/doc/book/pipeline/jenkinsfile/">Jenkinsfiles</a>)</li><li>Custom app monitoring in <a href="https://prometheus.io/">Prometheus</a>.</li><li>Logging to <a href="https://cloud.google.com/stackdriver/">Stackdriver</a>, using the <a href="https://docs.docker.com/engine/admin/logging/gcplogs/">Google Cloud Logging driver</a>.</li></ul><h3>Process for running new applications on the cluster</h3><p>The infrastructure team will become responsible for maintaining and improving the cluster. Their responsibility however, ends at the container level. All containers run in an orchestrator, so in a way they could affect each other. For example if one container starts dumping GBs of data on disk, it could affect disk performance or even cause an outage of the node.</p><p>At Google, applications need to be engineered so they are compatible with the <a href="https://research.google.com/pubs/pub43438.html">Borg cluster</a>. For example a service like the Multiplexer (responsible for starting up other R / Python / SQL containers) will not be able to run on our Kubernetes cluster out of the box.</p><p>This means that when there is a new service that should be deployed on the cluster, the engineers should sync with the infrastructure team, explain the high level functionality, expected load, expected CPU / RAM / Storage requirements, how downtime will affect other services, etc. Given <a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">part 1</a> on how to scale communication, this info is ideally part of the Github Wiki of the service.</p><h3>Which programming languages or frameworks do we use?</h3><p>Given that the choice has a big impact on recruiting, making cross-team pull-requests, speed for people to switch teams, etc. We decided to limit our stack for web applications to Ruby on Rails, NodeJS, React and Python.</p><p>Does this mean we will never use Go? No. However this does imply that if there is a team that would like to start using Go for a new service, it’s not something they should decide by themselves. Hence the idea is to start a broader discussion with the engineering team, to see if we are going to expand our stack to include Go.</p><p>Enforcing rules or limitations is never an easy task, however we hope that what we just described will increase productivity for all engineers in the long run, without limiting ownership.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cabde9c7a041" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/technical-vision-blog-part-3-cabde9c7a041">Technical Vision Blog — Part 3</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to scale DataCamp’s engineering group?]]></title>
            <link>https://medium.com/datacamp/how-to-scale-datacamps-engineering-group-f319a833bcd4?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/f319a833bcd4</guid>
            <category><![CDATA[company]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Jonathan Cornelissen]]></dc:creator>
            <pubDate>Sat, 12 Aug 2017 19:25:12 GMT</pubDate>
            <atom:updated>2017-08-14T17:24:29.892Z</atom:updated>
            <content:encoded><![CDATA[<p>This post focuses on how<em> to scale the people building the tech behind DataCamp, not on how to scale the actual technology</em> (for the latter, I refer to our <a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">CTO’s technical vision</a>). Scaling the engineering group in a smart way is essential for the company. The goal of this post is to lay the foundation for a structure that will allow DataCamp’s engineering group to double or triple in size in the next 12 months.</p><p><strong>DataCamp right now</strong></p><p>The simple fact that I’m writing this is a testament to the fact that we’re missing a key position on the DataCamp management team. We won’t be able to scale the team engineering team unless we have an end responsible for the <em>people</em> in the engineering group.</p><p>On the one hand, we want to grow the engineering group in a <em>highly decentralized fashion </em>with fairly independent teams, who can use different technologies, development practices, etc. (similar to <a href="https://vimeo.com/85490944">spotify</a>). On the other hand, we’ll have quite a few team leads and product managers over time, which will create a decent amount of chaos. Chaos is good and intended as innovation needs freedom and chaos. The remainder of this post focuses on how we can control that level of chaos and freedom. We need just enough people and process management to avoid total anarchy while maintaining innovation.</p><p><strong>Current problems</strong></p><p>Let’s start with a few examples of what’s currently not going well:</p><ul><li>2 of our engineering teams have become less productive in the last few months. No one addressed that.</li><li>There’s no one to coach new team leads on how to do their job well.</li><li>One team lost 2 people unexpectedly (which meant the entire team except the team lead was gone).</li><li>In the org chart all engineering team leads still report to the CTO, while the CTO wants to minimize time on people management, and <a href="https://medium.com/datacamp/cto-role-and-responsibilities-8b5d83733b29">spend more time on technical vision, architectural changes and building MVPs</a>.</li></ul><p>To be specific, the key problem is that there’s no clear end responsible for:</p><ul><li>People related performance issues (of team leads)</li><li>Speed/quality related performance issues (of teams)</li><li>Big projects that span multiple teams (e.g. gamification)</li><li>Structure of the engineering teams (when to split them etc.)</li><li>Engineering hiring flows: people will fit multiple teams often in the future with the micro-services architecture and someone needs to determine how these choices will be made.</li></ul><p>To sum all of that up: <br><em>While the technical vision is set by the CTO, someone else ultimately needs to own execution of the technical vision and hold people accountable to execute the technical vision. As the engineering team grows, ensuring the execution of the tech vision will become less straightforward, while it’s obvious with just a few engineers. And…</em></p><p>We’ll be growing teams and adding teams in the next few months. We’re moving to a situation with at least the following teams by our next company gathering in October:</p><ul><li>GE (Growth engineering — building all growth and marketing related initiatives)</li><li><strong>CE</strong> (Community engineering — building <a href="https://www.datacamp.com/community/">DataCamp’s community</a>)</li><li><strong>EE</strong> (Enterprise Engineering — building <a href="http://datacamp.com/groups">DataCamp’s groups product</a>)</li><li><strong>ME</strong> (Mobile Engineering — building our mobile app (launch Oct 16 ;-))</li><li><strong>LE</strong> (Learn Engineering — building the actual <a href="https://www.datacamp.com/courses">learning experience</a>)</li><li><strong>TE</strong> (Teach Engineering — building our <a href="http://datacamp.com/teach">course authoring environment</a>)</li><li><strong>EXE</strong> (Excel engineering — interactive learning for Excel #futuremusic)</li><li><strong>IE</strong> (Infrastructure engineering — building the infrastructure that makes all of the above run smoothly and gather data about it)</li></ul><p><strong>An assumption that’s no longer true…</strong></p><p>Given our focus on decentralization, ownership and microservices, we mistakenly assumed that close to zero management of engineering team leads and coordination between team leads was OK. The examples above &amp; future growth plans indicate that we’ve reached the stage where this is absolutely no longer true. <a href="https://medium.com/datacamp/cto-role-and-responsibilities-8b5d83733b29">The role of the CTO at DataCamp</a> has changed recently and I have moved away from regular check ins with engineering team leads some time ago, essentially leaving people &amp; project management of the engineering group to be owned by no-one.</p><p>By consequence, we now miss a position typically referred to as: VP engineering. Most companies create this role much earlier, our focus on ownership and decentralization has allowed us to postpone filling this role until now.</p><p><strong><em>So… what’s a VP engineering in a decentralized organization?</em></strong></p><blockquote><em>I have always compared them as that a VP of Engineering wakes up each morning concerned whether he/she has the absolutely best engineering team and the CTO wakes up concerned whether they have the absolute best technology.<br> — Werner Vogels — CTO Amazon</em></blockquote><blockquote><em>“A VP Engineering is ideally a great manager and a great team builder. He or she will be an excellent recruiter, a great communicator, and a great issue resolver. The VP Eng’s job is to make everyone in the engineering organization successful and he or she needs to fix the issues that are getting in the way of success.”</em></blockquote><p>For more context, please read:</p><ul><li><a href="http://avc.com/2011/10/vp-engineering-vs-cto/">CTO vs VP Engineering (AVC blog)</a></li><li><a href="https://medium.com/engineering-leadership/defining-roles-cto-and-or-vp-engineering-f1c7563643a3">Defining roles: CTO and/or VP Engineering</a> (from <a href="https://medium.com/u/f2cdd8176e0b">Sanford Dickert</a>)</li><li><a href="https://bothsidesofthetable.com/want-to-know-the-difference-between-a-cto-and-a-vp-engineering-4fc3750c596b">Want to Know the Difference Between a CTO and a VP Engineering?</a> (from <a href="https://medium.com/u/946f534320f7">Mark Suster</a>)</li><li><a href="https://www.ivyexec.com/executive-insights/2015/cto-versus-vp-engineering-whats-the-difference/">CTO vs VP Engineering: What’s the difference?</a></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/838/1*HyC_fIKhSrAZ6Bwi3fC5_A.png" /><figcaption>Image from <a href="https://bothsidesofthetable.com/want-to-know-the-difference-between-a-cto-and-a-vp-engineering-4fc3750c596b">Mark Suster’s blog on this topic</a></figcaption></figure><p><em>Both the CTO as well as the VP engineering should be part of the senior management team.</em> To make this very tangible: In senior management meetings, the CTO will answer questions and bring up concerns related to the technology (e.g. technical vision around microservices, what technologies should we use, what’s the feasibility of certain new projects, etc.). The VP engineering should be able to answer questions related to people and process of the engineering group (e.g. who’s ready to become a team lead, what’s the status of projects across teams, why have certain aspects of the tech vision not been implemented yet, how can we improve communication with product managers, etc.).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rIm8KPj7A-5hHEpmX2DWKA.png" /><figcaption>DataCamp’s engineering group (by end of October)</figcaption></figure><p>We believe scaling a decentralized engineering team will work best if we ensure that:</p><ol><li>Every <strong>Team Lead (TL)</strong> performs well and is held accountable: (i) the speed at which technology is being build (ii) the quality of the new tech.</li><li>There’s good communication between <strong>every TL</strong> and <strong>Product Manager (PM).</strong></li><li>There’s enough <strong>coordination</strong> between groups at the engineering level, while maintaining a high level of freedom at the team level. I think the default should be that we delegate all responsibilities to the team level, unless there’s a really good reason not to do so. For example: teams should be able to chose their own technologies, but having decent API documentation for example is not a choice but a clear requirement.</li><li>A good <strong>infrastructure team </strong>enables the entire engineering group.</li><li>We <strong>keep teams small </strong>(say max 4–6). This will help combat 2 challenges:<br>(i) We’re very open to remote work at DataCamp and think it’s an essential ingredient of our strategy to hire the best engineering talent. That said, communication becomes slightly harder when people are <strong>remote</strong>. It takes a really good people manager to manage a remote team of 10 people. Our current thinking is that it’s easier to avoid that situation altogether. As we want to be very flexible on remote work, the choice is to keep teams smalls. <br>(ii) A lot of (our) software engineers <em>don’t enjoy a lot of people management</em> (and it often doesn’t come natural to them, so there should be little risk of micro-management anyway). Keeping teams small ensures a manageable amount of “management” for every TL on the one hand. On the other hand, it makes it easy to evaluate if a team is still performing well (you can just look at the delta in features and code quality over a period of time for a few apps).</li><li>We hire engineers with product feeling who can manage the fact that they’ll get ownership and be the end responsible for one or more apps fairly fast. This will remain a big challenge as we haven’t found a good way of testing for it (except for internships). We should filter as much as possible for people who have a demonstrated ability to self-manage and build projects independently. Furthermore, we should clearly set expectations that if people can’t self-manage and take ownership of an app, DataCamp is not the right company for them.</li></ol><p><strong>The VP Engineering at DataCamp</strong></p><p>We’re very excited to welcome <a href="https://www.datacamp.com/instructors/filipsch">Filip Schouwenaars</a> in this new role. He’ll be transitioning into the new role in the next few weeks. At a high level, we envision his key responsibilities in the next 6–12 months to be the following:</p><ul><li><strong>Responsible for the engineering team (people)<br><em>- </em></strong><em>Attract new talent: <br></em>(i) Recruiting flows, coding tests, .. <br>(ii) Key point of contact for HR <br>(iii) Decisions in case people fit multiple teams<br><em>- Retention of talent: <br></em>(i) Make sure everyone is happy and aware of what their goals are <br>(ii) Weekly one-on-ones with team leads<br>(iii) Make sure people stay up to speed by going to conferences, etc. <br>(iv) Resolve communication issues with product managers <br><em>- Promote internal knowledge sharing: <br></em>For example: initiatives like lunch &amp; learn, etc.</li><li><strong>Responsible for the output of the engineering group<br></strong><em>- Make sure speed of execution remains high<br>- Balance engineering priorities with product management priorities<br>- Implementation of systems to ensure high code quality (testing, .. )<br>- Ensure good practices around documentation, internal communication, etc.</em></li><li><strong>End responsible for execution of projects across teams<br></strong><em>For example: Gamification, moving to a complete microservices architecture, …</em></li><li><strong>Representative of the engineering group at senior management level</strong></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f319a833bcd4" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/how-to-scale-datacamps-engineering-group-f319a833bcd4">How to scale DataCamp’s engineering group?</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Start with why]]></title>
            <link>https://medium.com/datacamp/start-with-why-bf0a07ab0ce4?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/bf0a07ab0ce4</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[company]]></category>
            <dc:creator><![CDATA[Jonathan Cornelissen]]></dc:creator>
            <pubDate>Sat, 12 Aug 2017 16:10:14 GMT</pubDate>
            <atom:updated>2019-07-26T16:46:00.676Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>Some context first on DataCamp</em></strong></p><p>At the start of 2016, DataCamp’s team consisted of 8 people. Less than 18 months later, we’ve grown to <a href="https://www.datacamp.com/about">45 people</a>. DataCamp’s subscriber numbers and courses have grown exponentially as well in that period. In the spirit of transparency, since the start of 2016:</p><ul><li># DataCamp subscribers grew from 4,900 to over 33,000</li><li># registered students grew from 260,000 to over 1.3 million</li><li># courses grew from 18 to 85</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*m4MyFpE-BMvupR6U." /></figure><p><strong><em>So why?</em></strong></p><p>While that’s incredibly exciting, it has forced us to rethink both internal as well as external communication flows. We hope to fix some of the internal communication challenges we’ve recently experienced by regularly explaining why DataCamp is doing certain things, by sharing mistakes, challenges, future goals, etc. In addition to improving internal communication, by being transparent, I hope other entrepreneurs may learn from our past and future mistakes in building a SAAS (education) company.</p><p>Two words of warning:</p><ol><li>Things change fast when a company is growing double digits month-over-month. In a few months, some of what gets written here now, may no longer be true or relevant.</li><li>I’m an <a href="https://en.wikipedia.org/wiki/INTP">INTP</a> and have a tendency to share <a href="http://avc.com/2016/02/intp/">undeveloped thoughts in a debate against myself</a>. I’ll try to share developed thoughts in these initial posts, but I’m pretty sure that won’t last long. A dream scenario would be that this becomes a scalable sounding board on what DataCamp is doing (and not doing).</li></ol><p>The initial posts will cover the following topics (among other things):</p><ul><li><a href="https://medium.com/datacamp/how-to-scale-datacamps-engineering-group-f319a833bcd4"><em>How to scale DataCamp’s engineering group?</em></a></li><li><a href="https://medium.com/datacamp/cto-role-and-responsibilities-8b5d83733b29"><em>CTO — Role and responsibilities at DataCamp</em></a></li><li><a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">Technical vision</a></li><li><a href="https://www.jonathancornelissen.com/post/why-we-started-datacamp"><em>Context on why we started DataCamp</em></a></li><li><a href="https://medium.com/datacamp/datacamp-the-plan-7ff04346fd6b">DataCamp: The plan</a></li><li><a href="https://www.jonathancornelissen.com/post/resources-for-first-time-ceo-s-and-founders-5-best-business-books-in-my-library">Resources for first-time CEO’s and founders — 5 best business books in my library</a></li><li><em>DataCamp’s culture</em></li><li><em>DataCamp’s funding strategy</em></li><li><em>…</em></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf0a07ab0ce4" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/start-with-why-bf0a07ab0ce4">Start with why</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Technical Vision — Part 2]]></title>
            <link>https://medium.com/datacamp/technical-vision-part-2-e3eb53dbaf5d?source=rss----428316015075---4</link>
            <guid isPermaLink="false">https://medium.com/p/e3eb53dbaf5d</guid>
            <category><![CDATA[company]]></category>
            <category><![CDATA[company-culture]]></category>
            <dc:creator><![CDATA[Dieter De Mesmaeker]]></dc:creator>
            <pubDate>Sat, 12 Aug 2017 15:55:03 GMT</pubDate>
            <atom:updated>2017-08-25T14:37:34.133Z</atom:updated>
            <content:encoded><![CDATA[<p>At DataCamp we have more than doubled our engineering team in the last 12 months. <a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">The first part</a> of this series goes in detail on which technical changes are required for us to scale our internal communication. This part goes more in depth on how we plan to change our technical architecture.</p><h3>Challenge 2: Scaling our architecture</h3><p>As mentioned in <a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">part 1</a>, we want engineers to take ownership over their services. This requires a significant amount of granularity in our architecture. (We don’t like having two teams both owning a service)</p><p>The fact that we have a “main app”, shows that we have too much logic in one place. This application is responsible for <a href="https://www.datacamp.com/groups/business">DataCamp for Groups</a>, integrations with other learning platforms, user authentication, course info, tracking user progress, etc. This has some implications:</p><ul><li>When the main app goes down, most of the other microservices go down.</li><li>If the Learn Engineering team wants to make a change to the API that the learning interface app uses, they need to make changes to the main app, which is the maintained by a different team.</li><li>There is no uniform system for authenticating users on different applications.</li><li>The Enterprise Engineering team depends too much on the capacity of the team maintaining the main app (Growth Engineering).</li></ul><p>Additionally, some services like the Projector app (responsible for video playback on DataCamp) are highly coupled to the main app. We want to have a clear separation of concerns between our services.</p><h3>Splitting the monolith</h3><p>Our flow for extracting the campus-backend (an API that will manage all data related to taking courses) from the main app would look something like this:</p><ol><li>Create a new campus-backend API in the main app. This API will interface exclusively with the database models we are extracting (Course, Chapter, Exercise, etc).</li><li>Split up all the logic in the main app, so it uses this internal campus-backend API, instead of calling the models directly.</li><li>Once all logic in the main app uses the new API the controller logic is split.</li><li>The new campus-backend should be well documented and end-to-end tested, so it can be refactored.</li><li>We duplicate the main app, deploy it, create a new DNS, but still let it connect to the same DB.</li><li>Update the endpoint in the main app to call the external DNS for accessing the campus-backend.</li><li>During a maintenance window, we split the database by promoting a read replica.</li><li>The Learn Engineering team will now become responsible for the new campus-backend service. They can choose if they want to refactor or rewrite it.</li></ol><h4>Separation of concerns</h4><p>In <a href="https://medium.com/datacamp/technical-vision-part-1-5f016c163340">part 1</a>, we talked about the benefits of looking at our internal services as if they were open source, with a focus on improving communication among teams. An additional benefit that we did not discuss, is that this mindset imposes questions like “<em>If we were to open source this service, would anyone use it?</em>”, “<em>How coupled is this library to our specific problem?</em>”, “<em>How hard would it be for other people to integrate with it?</em>”, etc.</p><p>Assuming that other teams will interact with the newly created service, as if it were to be an open source library, automatically puts you in a mindset of high decoupling.</p><p><a href="https://medium.com/datacamp/technical-vision-blog-part-3-cabde9c7a041">The next part</a> will go over how we want to enforce some consistency among our services, without limiting ownership in our engineering teams.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e3eb53dbaf5d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/datacamp/technical-vision-part-2-e3eb53dbaf5d">Technical Vision — Part 2</a> was originally published in <a href="https://medium.com/datacamp">DataCamp</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>