<?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[Stories by Nick Washburn on Medium]]></title>
        <description><![CDATA[Stories by Nick Washburn on Medium]]></description>
        <link>https://medium.com/@NickWashburn?source=rss-481411f3020c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*0WB9cZRqsXf0-nnip-gwNQ.jpeg</url>
            <title>Stories by Nick Washburn on Medium</title>
            <link>https://medium.com/@NickWashburn?source=rss-481411f3020c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 05:08:58 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@NickWashburn/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Secrets, Everywhere: Why We Co-Led Truffle Security’s Series B]]></title>
            <link>https://medium.com/@NickWashburn/secrets-everywhere-why-we-co-led-truffle-securitys-series-b-b4d5ef8337e8?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/b4d5ef8337e8</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[application-security]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[nhis]]></category>
            <category><![CDATA[iam-roles]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Thu, 06 Nov 2025 20:31:52 GMT</pubDate>
            <atom:updated>2025-11-06T20:31:52.785Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="Truffle Security team photo" src="https://cdn-images-1.medium.com/max/1024/1*g-o1_nUHE_RIe1VloIeG4A.jpeg" /><figcaption>Truffle Security Team</figcaption></figure><p><strong>By Nick Washburn &amp; Sunil Kurkure</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/truffle-blogsecrets-everywhere-why-we-co-led-truffle-securitys-series-b/"><em>Intel Capital website</em></a><em> on November 6, 2025]</em></p><p>Today, we are thrilled to announce that Intel Capital co-led <a href="https://trufflesecurity.com/">Truffle Security</a>’s $25 million <a href="https://www.axios.com/pro/enterprise-software-deals/2025/11/06/truffle-security-enterprise-credential-25-million">Series B financing</a> alongside Andreessen Horowitz.</p><p>In modern software development, secrets are everywhere. They’re the very bedrock that one machine or service uses to authenticate itself to another machine or service. These secrets — API keys, database passwords, private access tokens — power the interactions between thousands of services, libraries, and frameworks. The shift to microservices architectures, the decoupling of frontend and backends, and the explosion of third-party APIs have only accelerated this secrets proliferation — and it’s not slowing down. They’re hard to manage, hard to use, hard to understand: secrets are hard and they shouldn’t be.</p><p>Developers are increasingly relying on coding co-pilots and LLM-driven coding agents. While these tools promise productivity gains, they’re amplifying the secrets proliferation problem. First, these tools are rapidly accelerating the aggregate lines of code under management, which creates a much larger surface area to introduce code-specific vulnerabilities. Additionally, these tools are naturally non-deterministic and often make recommendations that don’t follow best practices for application security. A perfect example is hard coding secrets directly into the code base itself. Last year, the Truffle Security team tested ten popular LLMs and found that <a href="https://trufflesecurity.com/blog/llms-are-teaching-developers-to-hardcode-api-keys">most recommended hardcoding API keys and passwords directly in source code</a> — a ticking time bomb for application security teams. And in the rapid adoption cycle of AI, many developers are just accepting these recommendations at face value in the interest of speed and vibe development.</p><p>Secrets management platforms and storage systems have been around for nearly a decade, yet secrets continue to leak across Git repositories, messaging platforms like Slack and Teams, IT Service Management (ITSM) systems like Jira and Confluence, logs, API testing tools like Postman, and more. The result? Credential compromise remains the leading cause of data breaches, as highlighted in the <a href="https://www.verizon.com/business/resources/T7b6/reports/2024-dbir-data-breach-investigations-report.pdf">Verizon 2024 Data Breach Investigations Report</a>, often serving as the initial wedge for ransomware or extortion. While individual phishing exercises continue to be on the rise, many incidents are caused by inadvertent secrets leaks.</p><h4><strong>From Open Source to Enterprise Leadership</strong></h4><p>Truffle Security’s journey began with open-source projects that are now industry-wide standards leveraged across secrets detection. Dylan Ayrey, Co-Founder &amp; CEO, first created the open-source project <a href="https://github.com/trufflesecurity/trufflehog">TruffleHog</a> in 2016. Dustin Decker, Co-Founder &amp; CTO, and Dylan were classmates at Rochester Institute of Technology and have been core maintainers of the TruffleHog project for years. Additionally, the Truffle Security team includes Zach Rice, who first created the <a href="https://github.com/gitleaks/gitleaks">Gitleaks</a> open-source project in 2018 and joined Truffle Security in 2023, and Brad Larson, who first created the <a href="https://github.com/praetorian-inc/noseyparker">Nosey Parker</a> open-source project in 2022 and joined Truffle Security in 2025, extending Truffle Security’s open source advancement in the field by uniting core stewardship of all three projects under one roof. All three projects are extremely popular amongst practitioners, collectively with over 48K+ GitHub stars, 400+ contributors, and 15M+ DockerHub pulls.</p><p>Building on this foundation, Truffle Security is introducing <a href="https://trufflesecurity.com/trufflehog-analyze">TruffleHog Analyze</a>, which assesses cloud credential risk by clearly enumerating cloud resource access and permissions. This provides actionable insights to cloud security teams without exposing sensitive data. It is an extension of Truffle Security’s core product <a href="https://trufflesecurity.com/trufflehog-enterprise">TruffleHog Enterprise</a>, which provides secrets detection well beyond source code management tools to Slack, Jira, Confluence, Google Drive, and more, offering continuous monitoring, advanced permission analysis, key validations, and ticketing integration.</p><p>The team has designed the platform for enterprise readiness: SSO, encrypted data at rest, isolated environments, secure authentication, in-memory scanning, automatic updates, and third-party SOC 2 Type II audits ensure organizations can confidently deploy at scale.</p><h4>Why We Co-Led the Series B</h4><p>We were introduced to Dylan and Dustin in late 2021 and have since been following along on their journey as they expand from open-source secrets identification engines to broader identity and access management (IAM) workflows.</p><p>Secrets management is fundamentally a workflow problem for developers, not just a problem that can be solved by vaulting. Existing tools solve pieces of the puzzle, but breaches continue. Truffle Security’s approach uniquely positions them to tackle not just secrets identification but the broader IAM and non-human identity (NHI) and secrets lifecycle problem.</p><p>The growth of Truffle Security is a demonstration of the team’s technical excellence and relentless product mindset. They are very committed to supporting their large enterprise customers, constantly at the ready to rapidly implement new feature requests and build for security teams of the future. As a result, Truffle Security has seen strong commercial traction with large enterprises at very low burn ratios — a testament to the depth of their product and the ROI recognized by application security leaders.</p><p>Our investment accelerates Truffle Security’s mission to make secrets management frictionless, secure, and comprehensive for modern enterprises, and it positions them to tackle the broader IAM and NHI market opportunity with confidence.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b4d5ef8337e8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Moderne: The Antidote to Software Maintenance Toil]]></title>
            <link>https://medium.com/@NickWashburn/moderne-the-antidote-to-software-maintenance-toil-d221f777b0c9?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/d221f777b0c9</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software-maintenance]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[intel-capital]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Tue, 07 Feb 2023 17:47:03 GMT</pubDate>
            <atom:updated>2023-02-07T17:47:56.086Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="Moderne Co-Founders, Jonathan Schneider &amp; Olga Kundzich" src="https://cdn-images-1.medium.com/max/1024/1*TC91IAVThZ8munJz9CB3qQ.jpeg" /><figcaption>Moderne Co-Founders, Jonathan Schneider &amp; Olga Kundzich</figcaption></figure><p><strong>By Nick Washburn &amp; Alexandra Farmer</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/moderne-the-antidote-to-software-maintenance-toil/"><em>Intel Capital website</em></a><em> on February 7, 2023]</em></p><p>Stop for a moment and think about the house you live in. The structure was built using an endless supply of materials manufactured by third parties. From the drywall, to the plumbing, to the electrical wiring, all the way down to the cement and rebar comprising the very foundation. It’s a complicated structure built from many parts. And those parts degrade over time.</p><p>Now think about when you want to do an upgrade — say a new extension off the living room. You start the project, but quickly realize you have to totally replace the subfloor in the existing living room first, given it has degraded and is no longer up to spec. So before building new, you’re stuck maintaining and fixing the old.</p><p>Welcome to software maintenance…</p><h4><strong>Some Nightmarish Realities About Software Maintenance</strong></h4><p>The analogy above doesn’t come close to the complexity of software maintenance, but it is effective in illustrating a simple point: how can you build new when you can’t even maintain?</p><p>How did we get here? Software development has undergone massive changes in the past several decades, which ultimately led to our current state of affairs:</p><ul><li>Applications are assembled and less than 20% of the code base is custom business logic. 80% of the code base is third-party components.</li><li>The variety of those components — be it languages, frameworks, runtimes, libraries, APIs, webhooks to other services, etc. — is exponential.</li><li>The sheer volume (lines) of code has exploded. Organizations have substantially more lines of code and source repositories under management than they did 20 years ago.</li><li>Lastly, the rate of change for all of the above is… constant.</li></ul><p>The continuous compounding of the foregoing factors causes software maintenance to be the proverbial firehouse. The dependencies within your code base are changing frequently — shifting sands under your feet. Vast amounts of mission critical enterprise software are built in aging frameworks — introducing new vulnerabilities and bugs daily. Technical architectural choices made at one time may become burdens down the line — be it scaling challenges or otherwise.</p><p>The list goes on and on, but one thing is constant: Fixing these issues feels hopeless for most organizations. Enterprises have a massive backlog of maintenance activities that just sit there, creating further technical debt. All the meanwhile, engineering teams are being asked to ship new features and add business value. But 30%+ of their time goes to maintaining the old and existing.</p><h4><strong>You Can Tell Me I Have a Problem — But I Just Want It Fixed</strong></h4><p>For the past decade plus, there has been a tooling focus on identifying code problems, particularly focused on Common Vulnerabilities and Exposures (CVEs). The explosion of Static Application Security Testing and Software Composition Analysis tools have put a spotlight on demonstrating the breath of third-party dependencies in code bases. One challenge is the signal-to-noise ratio for those tools is pretty off the charts. Sure you can tell me I have a CVE in a repo, but if that code isn’t used (or used for anything critical) it doesn’t really impact my operations. So why would I prioritize it?</p><p>However, the bigger issue is these tools just tell you that you have a problem and nothing more. Engineering teams are then just left with manually remediating the necessary changes. So even if a code search or scanning tool identifies a CVE that is of critical importance, ultimately it comes down to engineering teams to (1) identify each repo, binary artifact and deployed asset where that CVE is contained, (2) manually refactor the code in the repos, and (3) build and ship the new into production as a new artifact and deployed asset. Burning a CVE down to zero in an org is highly manual.</p><p>All of this takes a ludicrous amount of time to do manually. To give a precise example, at one Moderne customer SonarQube estimated <strong>4.9 years</strong> of developer time to work to fix the top five identified CVE issues. Even small teams at SMEs have 20M+ lines of code. For large enterprises, that number is more like 500M+.</p><figure><img alt="Screenshot of Moderne Platform" src="https://cdn-images-1.medium.com/max/875/1*3zTNg6I_0wQ3kn6gzMezCQ.png" /><figcaption><em>In the screenshot above, the Estimated time savings of 413 hours comes directly from SonarQube’s report on these identified code issues. Moderne completed them in 3 hours, automatically.</em></figcaption></figure><p>There is also one critical component to refactoring beyond just the drowning toil of the overhead: style. Engineering organizations have conventions for how code is built within the company. Ensuring those conventions are maintained is key.</p><h4><strong>Automation Is Key — But Nothing Good Comes Easy</strong></h4><p>So, we’ve established this is a huge problem — how do you automate in an accurate way that developers and engineering leaders can trust? Unsurprisingly, it’s a very nuanced problem that requires significant engineering skill and a deep understanding of software languages. Fortunately, we were lucky enough to be introduced to the team that has solved it!</p><p>In June 2020, Casey Rosenthal, the CEO of <a href="https://www.verica.io/">Verica</a> and former Head of the Traffic &amp; Chaos Teams at Netflix, introduced Nick to Jonathan Schneider, formerly on the Netflix Engineering Tools Team before he went on to join the Spring Team at Pivotal and further on to Gradle Enterprise.</p><figure><img alt="Email introduction to Jonathan Schneider from Casey Rosenthal" src="https://cdn-images-1.medium.com/max/780/1*JPXlf7I5RWv3F98H2D-Z4Q.jpeg" /></figure><p>We quickly connected with Jonathan. He walked us through his experience building OpenRewrite (<a href="https://www.github.com/openrewrite/rewrite">repo</a>, <a href="https://www.docs.openrewrite.org/">docs</a>), which was proven at scale at Netflix. Jonathan had an incredibly in-depth, analytical framework on the needs for this solution. The ideation was that the high-level user objective is to promote change, to wipe away as much technical debt as possible to accelerate feature development. Importantly, he emphasized the product needs to focus on accuracy and accountability, which broadly requires:</p><ul><li>An asset inventory comprised on info on all assets across the software lifecycle (source code, binary artifacts, deployed assets) in order to understand current state and progress of a change campaign</li><li>The ability to fix code and dependencies, at scale, automatically</li></ul><p>Jonathan shared that he tweeted a teaser video of one auto-remediation in early 2020, and desperate enterprise Java developers zipped up their apps (source code!) and emailed them to Jonathan!</p><p>We left our first meeting incredibly impressed and set off to introduce Jonathan to potential design partners for the product, while Jonathan spent time looking for the perfect co-founder — Olga Kundzich. Olga and Jonathan worked together at Pivotal, where Olga was the product lead for Spinnaker. She was also heavily involved in data protection at EMC and the perfect complement to Jonathan. In addition to their impressive technical depth, Olga and Jonathan bring intense understanding and empathy to customers, as they feel compelled to fix this problem for them.</p><h4><strong>Moderne: Continuous Software Modernization</strong></h4><p>During the course of the next year and a half, we stayed in regular contact with Jonathan and Olga as the company was incorporated, the product was built, and design partner customers signed up. The feedback from the engineering leaders we introduced them to in large, enterprise companies was honestly pretty incredible and something we rarely see in tooling that early in its development. As we prepared this blog, we went back to our notes from some of those early calls, when the product was still in its infancy (it’s SOC2, Type 2 now and deployed at scale) just to ensure we weren’t having confirmation bias now!</p><p>And more importantly, the ROI that some enterprises using the early product were achieving was nothing short of magical. One dev team of 12 people was responsible for 420 source code repositories comprising 100s of millions of lines of code. They spoke about a framework migration that had been underway for over two years. Using Moderne, a single engineer was able to complete this migration in a day!</p><p>Of course, we loved the fact that the team was extremely careful to ensure there was no change to the developer’s day-to-day workflows. The Moderne platform simply plugs-in to existing build tooling and you’re off.</p><p>Pat Johnson, Director of Product Marketing at Moderne, wrote a wonderful overview of <a href="https://www.moderne.io/blog/continuous-software-modernization-with-moderne">Continuous Software Modernization</a> that we’d encourage you to read. It clearly outlines the use cases, how the tooling compares to code scan and search tools, and provides clear, actionable updates and migrations to start with. It also provides a glimpse into the <a href="https://www.moderne.io/blog/lossless-semantic-trees-the-key-to-restoring-creativity-to-software-engineering">Lossless Sematic Tree</a>, which is a unique data model that underpins the ability to do highly performant search and — most importantly — repeatable, automation-driven refactoring and remediation at scale with minimal developer disruption.</p><p>Moderne’s platform supports Java and is quickly extending support for additional languages like COBOL (released in December 2022), Python, TypeScript, and C#. In addition to being SOC 2, Type 2 <a href="https://www.moderne.io/blog/moderne-achieves-soc-2-type-2-certification">certified</a>, the team is working on additional capabilities to operationalize continuous software modernization at scale. These include change campaign management, a recommendation service, a CLI to drive the local developer experience, and integrations with CI tools, IDEs, and other tools in the ecosystem.</p><p>Today, we’re thrilled to announce leading <a href="https://www.techcrunch.com/2023/02/07/moderne-is-bringing-automated-code-remediation-across-complex-code-bases/?guccounter=1&amp;guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&amp;guce_referrer_sig=AQAAABXqbYurH_rJZIrjzq7u51-NaIvER5GFkpO4SKMLVtCq9EugATHhMz5iNGP1Ht6WrTbshwQXF-ofksoDFgrjVneAp43sULSKhT1ingtWBn8PuL_3SvdXvkoqXBMYKv75drzFXqNxoM2p999OnhF1CRiSh_DNyy49zr5Az3gWzmxo"><strong>Moderne’s $15M Series A round</strong></a>, together with our friends at True Ventures, Mango Capital, and Allstate Strategic Investors. This problem is too big to not be solved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d221f777b0c9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[2023 Areas of Focus]]></title>
            <link>https://medium.com/@NickWashburn/2023-areas-of-focus-1899b4ae544f?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/1899b4ae544f</guid>
            <category><![CDATA[developer-tooling]]></category>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[developer]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[infrastructure-software]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Tue, 17 Jan 2023 04:41:47 GMT</pubDate>
            <atom:updated>2023-01-17T04:46:36.041Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SxiQqAnVxPwHmaZ6bOIQBQ.png" /><figcaption>2023 Road To Travel</figcaption></figure><p><strong>By Nick Washburn</strong></p><p>For 2023, here are some interesting areas within cloud infrastructure and developer tooling that I’m particularly focused on from a potential investment perspective. This list is not intended to be exhaustive, and similarly there are many ways to categorize these areas. If you are building an early-stage company or otherwise working on these problem statements, I’d love to chat, learn more, and find ways to help:</p><ol><li><strong>WebAssembly Outside of the Browser (Particularly Tooling Around Debugging/Observability) </strong>— I spent a good portion of 2022 learning as much as I could about <a href="https://webassembly.org/">WebAssembly</a> (WASM), the <a href="https://wasi.dev/">WebAssembly System Interface</a> (WASI), the component model, and how teams are currently using WASM outside of the browser in the wild. While much ink has been poured on the benefits of WASM outside of the browser (isolation, extensibility, performance to name a few), it is admittedly still very early days on the first/best use cases. Nevertheless, it’s an exciting space to watch and I’ve been fortunate to meet with a lot of great early-stage teams (including <a href="https://cosmonic.com/">Cosmonic</a>, <a href="https://dylib.so/">Dylibso</a>, <a href="https://www.fermyon.com/">Fermyon</a>, <a href="https://suborbital.dev/">Suborbital</a>, <a href="https://www.secondstate.io/">Second State</a>, etc.) as well as large industry players like Intel itself. One constant feedback I’ve heard is a ‘here and now’ problem is the developer tooling — particularly around debugging. Every conversation around ‘what needs to be solved right now’ has brought this up. So it’s an area I’m particularly focused on, as regardless of the use cases for WASM outside of the browser, it’s clear it needs to be debugged and observed. The Dylibso team is doing cool stuff here, but I’m open to meeting with anyone else in the space working on this problem.</li><li><strong>The Magic of Control Planes</strong> — Control planes are magical, and where the Cloud Service Providers (CSPs) derive a lot of their value. Whether it is a third-party, opinionated PaaS, your own self-built internal developer platform, or just the inner-workings of a standard, multi-tenant SaaS application, control planes are everywhere. I’ve had a front row seat to seeing the benefits of control planes by working with the team at <a href="https://www.upbound.io/">Upbound</a> and witnessing the adoption of the <a href="https://www.crossplane.io/">Crossplane</a> project. However, there is certainly more to be down with control planes. For instance, whether it’s multi-tenant SaaS applications or API-first developers, there is just so much ‘recreating the wheel’ over and over. I feel that common control planes can be used to abstract away common developer stacks that underpin a lot of full-stack applications. There are great teams working on solving this in addition to Upbound, including <a href="https://www.katanemo.com/">Katanemo</a> and <a href="https://www.thenile.dev/">Nile</a> to name a few. It’s an exciting space I’m tracking closely.</li><li><strong>Works On My Machine, Not In Production</strong> — In this case, I’m starting with a problem statement, rather than a specific area of technology. And this problem statement is (1) still huge and (2) not close to solved. I’m particularly excited about this problem given the countless angles to attack it. One way to skin this problem statement is dependency and package management. Great teams working in this space include the <a href="https://floxdev.com/">flox</a> team around the <a href="https://nixos.org/">Nix</a> project, the <a href="https://www.tangram.dev/">Tangram</a> team, etc. In addition, the environments themselves are another way to focus on this challenge. There are a host of companies tackling development environments from a variety of angles, including <a href="https://www.raftt.io/">Raftt</a>, <a href="https://release.com/">Release</a>, <a href="https://replit.com/">Replit</a>, <a href="https://roost.ai/">Roost</a>, <a href="https://www.usenimbus.com/">Nimbus</a>, <a href="https://stackblitz.com/">Stackblitz</a>, etc. Similarly adopting the benefits of monorepos are another avenue towards this overall problem. <a href="https://monorepo.tools/">Monorepo Tools</a> is a great site that was developed by the team at <a href="https://nx.app/">Nx</a> in conjunction by a number of industry players around popular build and monorepo tools. Last but not least, code mapping and dynamic runtime analysis — like <a href="https://appmap.io/">AppMap</a> and <a href="https://www.codesee.io/">CodeSee</a> — offers great visibility into understanding your dependency footprint. Just given the breath of this overall problem statement, I’m spending a good amount of time here learning.</li><li><strong>Modern Distributed Cloud Stack</strong> — I’ve previously written about this in connection with our Series A investment in <a href="https://fly.io/">Fly.io</a> (<a href="https://medium.com/@NickWashburn/the-compute-layer-for-the-modern-distributed-cloud-our-investments-in-fly-io-bd171eda60b2">The Compute Layer for the Modern, Distributed Cloud</a>), but I believe a modern, distributed cloud stack is emerging. It must have a wonderful developer experience at each layer, which is something the large CSPs and CDNs sorely lack). While I believe Fly will become the intelligent, distributed compute in this stack, I’m interested in teams tackling other layers. This includes front-end workflows and state management, overlay networking and, ultimately, the databases behind it all.</li></ol><p>While the areas above are not exhaustive, they do represent targeted domains I’m spending more time in this year.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1899b4ae544f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Freshpaint: A Fresh Approach to Customer Data]]></title>
            <link>https://medium.com/@NickWashburn/freshpaint-a-fresh-approach-to-customer-data-86b81d948ebd?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/86b81d948ebd</guid>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[product-analytics]]></category>
            <category><![CDATA[customer-data-analytics]]></category>
            <category><![CDATA[data-management-platform]]></category>
            <category><![CDATA[customer-data]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Wed, 16 Nov 2022 20:05:51 GMT</pubDate>
            <atom:updated>2022-11-16T20:05:51.001Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n2UEzV6LXS6M6oFiLvjVKg.jpeg" /><figcaption>Freshpaint Co-Founders: Michael Malis &amp; Steven Fitzsimmons</figcaption></figure><p><strong>By Nick Washburn, David Mueller &amp; Alexandra Farmer</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/freshpaint-a-fresh-approach-to-customer-data/"><em>Intel Capital website</em></a><em> on November 16, 2022]</em></p><p>Inefficiencies have long offered a great opportunity for innovation. Iconic companies have been built by founders who just looked at a status quo process (often then called “the standard”), scratched their heads, and were impelled to fix it. Never satisfied by “that’s just how it’s been done”, the itch to fix a universal problem is just too compelling.</p><p>We’ve felt this itch too at Intel Capital, particularly looking for founders fixing huge inefficiencies within the enterprise data circulatory system — a vast and complex beast with many stakeholders and processes. And when looking at enterprise data, the crown jewels are self-generated, first-party customer data.</p><h4>First-Party Customer Data is the Center of Businesses</h4><p>Companies of all sizes run their businesses with customer data. It’s the lifeblood of a business. From multinational enterprises to early-stage startups, each of these companies must measure the needs of their customers and how they interact with their products. That vital information informs… everything.</p><p>Customer data is universally the best input for product roadmaps, marketing campaigns, sales calls, customer success engagements, growth hacks — the list goes on and on. And naturally the value of that customer data increases when (1) it is proprietary and generated by your own customers, (2) it is consumable by those that need it, (3) it is timely and actionable, (4) and it can be enriched.</p><p>As customer behaviors become more dynamic, the pressure is on for product, marketing, growth, and sales teams to understand customer signal and inform new features that drive product discovery, activation, and increased consumption by customers. This in turn drives sales and expansion. A virtuous flywheel powering the business.</p><h4>The Status Quo of Customer Data Collection is Sadly Inefficient Handoffs</h4><p>Despite the fact that the first-party customer data is used and owned by non-engineers — be it product, marketing, growth, or sales teams — engineering has historically owned the instrumentation and collection of that data. Specifically, engineering teams had to stand up precision tracking code, which is very tedious work. Maintaining that tracking code is also extremely laborious, particularly when what you need to actually measure changes day to day or even minute to minute. New log-in flow: measure it. New UI on the check-out page: measure it. Alternative marketing campaigns in the North East region on Labor Day: measure it. You get the drift.</p><p>So companies have been stuck in this trap where non-engineers need first-party customer data — and need it in real-time — yet they spin up Jira tickets for engineering teams to adjust precision tracking code that ends up just getting backlogged because those same engineers also need to focus on new features, product stability, and countless other workstreams. And this process results in product, marketing, and growth leaders waiting a month or more every time they need engineers to set up new customer events. This is crippling their biggest strategic advantage: Speed.</p><h4>Enter a Fresh Approach to Customer Data Collection</h4><p>When our good friend Taylor Clauson at Abstraction Capital reached out about connecting with the <a href="https://www.freshpaint.io/">Freshpaint</a> team, we were immediately struck by our thesis fit. Freshpaint puts the power directly into the hands of non-engineers, allowing product, sales, marketing, and growth teams to directly and automatically capture first-party customer data and consume it within the tools they already use day to day.</p><p>Once a single JavaScript SDK is installed, the Freshpaint platform automatically starts to capture a multitude of customer interactions — be it clicks, pageviews, form submissions, or the like. This automatic, Day 1 collection is huge in that data consumer teams don’t always know — right now — what they are hoping to measure. After all, product experiments evolve over time. Behind the scenes, Freshpaint is automatically collecting all of that information and exporting it directly into your cloud data warehouse, which is the source of truth for all company data. Months later, if your growth team wants to understand how a specific feature roll-out caused impact to net revenue retention, that data is magically there!</p><p>Freshpaint’s <a href="https://www.documentation.freshpaint.io/data-management/overview-of-features">features</a> are not limited to auto-track though. There are countless data <a href="https://documentation.freshpaint.io/data-management/sources">sources</a> supported (even precision tracking if you really want to!), and very powerful data <a href="https://documentation.freshpaint.io/data-management/transformations">transformations</a> are supported. And of course, the list of <a href="https://www.freshpaint.io/integrations">native integrations</a> for eventual data sources put this captured and transformed data directly into the tooling non-engineers use day to day. Freshpaint even offers HIPAA <a href="https://documentation.freshpaint.io/data-management/hipaa-mode">compliant</a> features! As you read through the documentation, it should become clear this is a very powerful platform — ultimately empowering non-engineers with a streamlined workflow. In addition to Freshpaint’s stellar product, we were equally impressed by the founders behind it. Co-Founders Fitz and Malis have a powerhouse of domain knowledge stemming from their experiences building and scaling customer analytics infrastructure. We fortunately had the opportunity to partner with this great team, building a best-in-class product, with impressively efficient growth metrics, by leading their <a href="https://venturebeat.com/data-infrastructure/looking-to-ease-engineers-from-customer-data-tasks-freshpaint-raises-14-5-million/">Series A</a>. We look forward to joining their journey helping non-engineering teams drive their most important customer-related decisions.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=86b81d948ebd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Compute Layer for the Modern, Distributed Cloud: Our Investments in Fly.io]]></title>
            <link>https://medium.com/@NickWashburn/the-compute-layer-for-the-modern-distributed-cloud-our-investments-in-fly-io-bd171eda60b2?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/bd171eda60b2</guid>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[back-end-development]]></category>
            <category><![CDATA[developer]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[full-stack]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Thu, 28 Jul 2022 16:41:24 GMT</pubDate>
            <atom:updated>2022-07-28T16:41:24.212Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="Two balloons looking at the Earth from the Moon" src="https://cdn-images-1.medium.com/max/1024/1*5wJkEx7ylUa3WzdGRbrqWA.jpeg" /><figcaption>Fly Me To The Moon</figcaption></figure><p><strong>By Nick Washburn &amp; Alexandra Farmer</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/the-compute-layer-for-the-modern-distributed-cloud-our-investments-in-fly-io/"><em>Intel Capital website</em></a><em> on July 28, 2022]</em></p><p>It all began with a simple question we asked ourselves:</p><blockquote><em>If you were a full-stack developer and could take an existing application, package it in a Docker image, and distribute it as close to your end-users as possible using a wonderful developer experience without any refactoring, why wouldn’t you?</em></blockquote><p>This question was prompted by an extremely popular <a href="https://news.ycombinator.com/item?id=22616857">thread</a> we read in March 2020 on Hacker News upon the official launch of the <a href="https://fly.io/docs">Fly Global Application Platform</a>.</p><p>As we digested the lengthy conversation in the thread, we kept centering on this golden question. It just seemed obvious to us that this was the way applications needed to be distributed. So we set off to stress test our assumptions and look for reasons that this shouldn’t just become… <em>the standard</em>.</p><h4><strong>To the Edge and Beyond!</strong></h4><p>Cloud architecture continues to push closer to the edge and end-users, as latency fundamentally impacts user experience. Users love “instant” applications. Think about the last time you used an application that lagged — it’s an annoying experience. Sadly due to pesky things like the speed of light, you can’t really deliver an application “instantly.” But for a full-stack application, 100 milliseconds (ms) is the magic threshold. Response times below 100ms <em>feel</em> instantaneous to end-users. 40ms all the better!</p><p>Historically, the Content Delivery Networks (CDNs) have done a good job leveraging locational caching for static content. Innovation within the front-end community around the JAMStack (JavaScript + APIs + Markup) has similarly greatly simplified front-end delivery (e.g., Netlify and Vercel). Hosting and rendering that static content in a point of presence (PoP) close to an end-user helps drive this “instant” feeling.</p><p>However, the back-end/full-stack engineering community had yet to really see a similar platform because the intelligent routing of application logic — including dynamic content and online transactions– to end-users at these thresholds is extremely complex for most developers. Not to mention terribly costly if done improperly.</p><p>But the market dynamics are there: users expect fast applications and happy users drive better business metrics. Whether it is the CDN market segment <a href="https://www.marketsandmarkets.com/Market-Reports/content-delivery-networks-cdn-market-657.html">forecasted</a> to grow from $14.4B in 2020 to $27.9B in 2025, at a CAGR of 14.1%, or the edge computing market segment <a href="https://www.marketsandmarkets.com/Market-Reports/edge-computing-market-133384090.html">forecasted</a> to grow from $3.6B in 2020 to $15.7B in 2024, at a CAGR of 34.1%, or the serverless architecture market segment <a href="https://www.marketsandmarkets.com/Market-Reports/serverless-architecture-market-64917099.html">forecasted</a> to grow from $7.6B in 2020 to $21.1B in 2025, at a CAGR of 22.7% — any way you slice it the market is pulling compute closer to the end-user.</p><h4><strong>You Can Do Magic!</strong></h4><p>So back to our original question. As we read more about Fly’s platform, it was clear the team created an amazing and intuitive developer experience, particularly when compared to the status quo.</p><p>Modern applications are commonly packaged into containers and capable of carrying all the code needed to operate. Sadly, they are often deployed into specific datacenters and never moved. This ends up in a vicious cycle of globally duplicating resources in every datacenter near(ish) end-users — which is the furthest thing from sustainable.</p><p>Think about a shipping container — they are built to a universal standard and intentionally meant to be transported to where needed. This was the Fly platform in a nutshell -&gt; moving compute around the world to where the demand is without additional headaches.</p><p>The Fly <a href="https://fly.io/docs/reference/architecture/">architecture</a> is highly abstracted and on a unified architecture leveraging bare-metal servers in established <a href="https://fly.io/docs/reference/regions/">regions</a>. From <a href="https://fly.io/docs/flyctl/">flyctl</a> (a CLI tool installed into local machine), a developer simply needs to add 4 lines of code. There is no refactoring of any application logic. So long as the code can be put in Docker image, it can leverage Fly.</p><p>When the developer ships the app to Fly, it is deployed on a very lightweight <a href="https://github.com/firecracker-microvm/firecracker">Firecracker microVM</a> to a number of locations (typically starting with the developer’s location), each connected via <a href="https://www.wireguard.com/">WireGuard</a> (a networking security mesh), and is given a singular global IP address via <a href="https://fly.io/docs/reference/architecture/#bgp-anycast">BGP Anycast</a>. By default, Fly listens for HTTP and HTTPS connections, but can be configured for any type of TCP traffic.</p><p>Then when traffic comes in, the Fly platform can identify the location via IP address and dynamically assigns compute resources in datacenters closest to that traffic.</p><p>So stepping way back, a back-end/full-stack developer can take an existing application, add 4 lines of code, and from the CLI globally distribute the application to scale closer to end-users. It just seemed… <em>magical</em>.</p><h4><strong>Fly Me To the Moon</strong></h4><p>As we continued to monitor Fly’s progress from afar throughout 2020 and early 2021, we quickly realized we were not the only folks that were very jazzed about what the Fly team was doing. Whatever internet rabbit hole we went down, we would find early users of Fly that <em>loved</em> the platform. (Sidebar: this love continues to this day.) Fortunately, one of our good friends Daniel Doctor at Dell Technologies Capital had participated in the Fly seed round and graciously introduced us to Kurt Mackey, Co-Founder and CEO.</p><p>We immediately hit it off with Kurt with a shared vision. The individual developer was key, as the developer experience on the Fly platform was critical. Kurt was extremely smart about listening to the needs of various full-stack communities and helping them solve business challenges. For instance, take Fly’s love for <a href="https://fly.io/blog/we-are-hiring-elixir-developer-advocates/">Elixir</a> -&gt; which led to Chris McCord <a href="https://twitter.com/chris_mccord/status/1428821074553315332">joining Fly</a> and the launch of the <a href="https://fly.io/phoenix-files/">Phoenix Files</a>.</p><p>In addition, the remainder of the Fly founding team was incredible. Together, Kurt, Jerome, Michael and Thomas not only built an incredibly sophisticated platform that users loved, they had also established a world-class company culture driven by compassion and deep empathy for the end-developer.</p><p>Couple that with the explosive growth we were witnessing in the platform since launch, we <em>had</em> to partner with Fly. As a result, we subsequently led the Series A round in August 2021 and got to work.</p><p>Since that time, the continued growth and innovation within the Fly platform has been incredible to watch. Whether it was the launch of <a href="https://fly.io/blog/free-postgres/">Free Postgres</a>, Fly’s acquisition of the <a href="https://fly.io/blog/all-in-on-sqlite-litestream/">Litestream project</a>, or the availability of <a href="https://fly.io/blog/fly-machines/">Fly Machines</a> for building your own FaaS service, the hits keep on coming! It is amazing how a community behind a wonderful product can help drive innovation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/730/1*bGGE5OnVHVWeUbZwxFTUAw.png" /><figcaption>Community-Led-Product-Led Growth!</figcaption></figure><p>Beyond that, we continue to have a deeply held belief that the modern, distributed cloud is quickly emerging. It is table-stakes that it will have a wonderful developer experience at each layer (something the large CSPs and CDNs sorely lack). And center to this new cloud stack is intelligent, distributed compute -&gt; where we believe Fly will become the <em>standard</em>.</p><p>Congratulations to the Fly team on <a href="https://techcrunch.com/2022/07/28/fly-io-wants-to-change-the-way-companies-deploy-apps-at-the-edge/">today’s joint announcement of our original $12M lead Series A round, as well as the subsequent $25M Series B round led by a16z</a>! Back to building!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bd171eda60b2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Database Optimization on Autopilot: Our Investment in OtterTune]]></title>
            <link>https://medium.com/@NickWashburn/database-optimization-on-autopilot-our-investment-in-ottertune-9c8f7b5bd22e?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/9c8f7b5bd22e</guid>
            <category><![CDATA[database]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[optimization-software]]></category>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[database-administration]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Wed, 11 May 2022 18:26:49 GMT</pubDate>
            <atom:updated>2022-05-11T18:26:49.407Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="OtterTune Co-Founders" src="https://cdn-images-1.medium.com/max/1024/1*qGloEsdLBjEGZzpUaJJYjQ.jpeg" /><figcaption>OtterTune Co-Founders, Andy Pavlo, Dana Van Aken and Bohan Zhang</figcaption></figure><p><strong>By Assaf Araki &amp; Nick Washburn</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/database-optimization-on-autopilot-our-investment-in-ottertune/"><em>Intel Capital website</em></a><em> on May 10, 2022]</em></p><p>The database is a complex beast. To optimally install and configure a database, one needs to understand how to choose the right CPU, memory, storage system, and network. There will be a different software (SW) configuration for every hardware (HW) setting, including the storage files, tables definition, and indexing. The goals of the SW and HW configurations are to answer the needs of the queries that will run on the database management system (DBMS). Once the DBMS goes live, it changes as application workloads change and evolve at runtime — and the configurations need to track the workloads and usage to stay optimized.</p><h4><strong>DBMS Trends During the Last Decade</strong></h4><p>The last decade has witnessed a number of clear trends with respect to DBMS adoption:</p><ul><li><strong>Overall, DBMS usage and market size has grown rapidly and continues to do so.</strong> For example, Gartner estimated in their annual <em>Market Share </em>reports that the DBMS market was $23.3 billion in 2010 and it <a href="https://www.gartner.com/en/documents/4001330">reached $65.8 billion</a> in 2020.</li><li><strong>The share of open-source DBMSs has rapidly increased during that period. </strong>In their <a href="https://www.gartner.com/en/documents/3970418">State of the Open-Source DBMS Market, 2019</a> report, Gartner estimated that by 2022 more than 70% of new in-house applications would be developed on an open-source DBMS or an open-source database-as-a-service (DBaaS) offering. They further estimated that 50% of existing proprietary relational DBMS instances will have been converted or be in the process of converting by that time. DB-Engines ranks DBMS using a range of data sources, and as of May 2022, open-source DBMSs surpassed commercial licensed DBMSs, <a href="https://db-engines.com/en/ranking_osvsc">51.4% to 48.6%</a>.</li><li><strong>Relational DBMSs continue to represent the lion’s share of the most popular databases.</strong> Gartner estimated in their <em>Market Share</em> report that relational DBMSs captured <a href="https://www.gartner.com/en/documents/4001330">83% of the DBMS market in 2020</a>. Further illustrating the point, as of May 2022 relational DBMSs continue to represent the <a href="https://db-engines.com/en/ranking_categories">most popular category at 72%</a> on DB-Engines rankings.</li><li><strong>Cloud relational DBMSs have grown the fastest of them all. </strong>In October 2009, Amazon Web Services (AWS) released Relational Database Service (RDS), supporting MySQL and later expanding to a variety of other relational databases, including PostgreSQL, SQL Server, and Oracle. A <a href="https://d1.awsstatic.com/whitepapers/RDS/Stack-Overflow-The-Incredible-Growth-of-Amazon-RDS.pdf">study from Stack Overflow</a> revealed that traffic to discussion pages about RDS increased 40% year-over-year from 2017 to 2018. Further, Valuates Reports estimates that the cloud DBMS market segment will <a href="https://www.bloomberg.com/press-releases/2021-07-20/cloud-database-market-to-reach-usd-68720-million-by-2026-at-a-cagr-of-38-2-valuates-reports">reach $68.7 billion by 2026, at a CAGR of 38.2%</a>.</li></ul><p>Stepping back, the past decade clearly saw incredible growth in usage and adoption of relational DBMSs — particularly cloud DBMSs based on open-source projects. On one hand, this growth largely made DBMSs more accessible to non-database professionals. Cloud DBaaS offerings from the major cloud-service providers (CSPs) carried high-levels of abstraction. As a result, SW developers and data analysts were able to deploy relational databases with ease.</p><p>On the other hand, underneath the hood something else was also increasing: complexity. During this period, the number of system knobs available for configuration and tuning increased dramatically. By some estimates, configuration knobs on relational DBMSs have grown 5–7X in the last twenty years.</p><p>So we were left with this interesting juxtaposition — more users were adopting and deploying relational databases with ease, while the underlying operational complexity skyrocketed. And worse, the “experts” at running DBMSs at scale — database administrators (DBAs) — were not growing in parallel. According to the <em>2020 Data Management Staffing Ratios report</em>, IT staff grew only by 1% in the last two years. A typical enterprise organization has 100s to 1000s of DBMS instances. Yet, human DBAs were minimal and could only monitor and operationally support a fraction of them.</p><p>Lastly, during the past decade the “Ops” businesses have grown rapidly, from ITOps solutions that monitor SW and HW, to FinOps solutions that focus on cost monitoring, to MLOps that manage ML models. Nevertheless, most database optimization and tuning services still were handled by consulting companies who manually reviewed the state of the DBMS occasionally and then manually tuned it. If a company had enough optimization or resiliency needs, they may then have hired a full-time DBA. However, even the best DBA cannot go over the endless configuration options to optimize a DBMS and they certainly are not available for the job 24x7.</p><h4><strong>A New Approach Is Needed &amp; Born Out of CMU Research</strong></h4><p><a href="https://ottertune.com/">OtterTune</a> was founded in 2020 by Andy Pavlo, Dana Van Aken and Bohan Zhang, but the idea of the concept originated many years prior stemming from academic interests of Andy. During Andy’s original Ph.D., he worked on automated methods to speed up transaction processing databases. When Andy then joined Carnegie Mellon University (CMU) as a professor in 2013, he continued to think about further automated methods for improving databases.</p><p>The “ah-ha” moment for Andy stemmed right after joining CMU in fact. During a visit to a large financial institution to explore industry problems around database operations, Andy’s eyes were opened to the amount of money the company was spending on what were — effectively — very basic maintenance tasks. This meeting spurred Andy to go deeper into his research on autonomous databases.</p><p>The timing was very critical, as in parallel there was a resurgence in machine learning (ML) studies at that time on account of readily available datasets, effectively “infinite” compute in the cloud, increased HW acceleration in that compute for ML workflows, and new open-source ML frameworks. Accordingly, Andy was spurred by the recognition of a clear industry problem and a tipping point in available ML technologies. He wanted to dive deep into applying ML towards autonomous database optimizations.</p><p>Together with Dana and Bohan, who were then students of Andy at CMU, Andy started the OtterTune <a href="https://db.cs.cmu.edu/projects/ottertune/">research project</a> at CMU in 2014. Their original thesis was to use ML to autonomously optimize DBMS knob configurations. As mentioned above, this was during the period that there was an explosion in knob complexity within DBMS — leading to a very difficult area for DBAs to configure in practice. Ultimately, Dana’s <a href="https://db.cs.cmu.edu/papers/2021/vanaken-thesis2021.pdf">Ph.D. dissertation</a> was the result of this research at CMU, with Dana leading the design and development of the original prototype.</p><p>Andy, Dana and Bohan published their <a href="https://db.cs.cmu.edu/papers/2017/p1009-van-aken.pdf">work</a> in SIGMOD in 2017, and it was well received by the academic community. Fortunately, the original SIGMOD paper had been funded by a research grant from Amazon, and the Amazon team invited Andy and Dana to write a <a href="https://aws.amazon.com/blogs/machine-learning/tuning-your-dbms-automatically-with-machine-learning/">blog article</a> about the product. The blog resulted in Dana being inundated with emails from people asking her to run OtterTune to optimize their databases.</p><p>Sensing this very strong market signal — when coupled with all of the notable trends the past decade outlined above — Andy, Dana and Bohan (who had since graduated at CMU) decided to incorporate OtterTune. From our perspective, Andy, Dana and Bohan are the perfect team to tackle this problem, as it is highly technical and multi-disciplinary at the intersection of database field and applied ML.</p><h4><strong>OtterTune Brings a Safety Net to Let Users Focus on Data and Insight — Rather than Worrying About the Database</strong></h4><p>Since incorporating and launching OtterTune in 2020, the team has released a highly accessible optimization platform that has expanded well beyond knob tuning, which is only one of many aspects of the database lifecycle. Supporting databases on Amazon RDS and Aurora, OtterTune’s platform automatically ensures that these databases run with the proper configuration settings, query plans, indexes, and replication schemes. Beyond just performance improvements and cost reductions, OtterTune focuses on OLTP DBMSs that are the backbone of any company operation, critical for the business, and provides a safety net for running cloud databases — freeing up developer and DBA time to focus on more impactful tasks.</p><p>Similarly, launching the OtterTune team has further expanded the product to cover health checks that identify specific database problems — automatically finding the proverbial “needle in the haystack.” Many developers understand that something in a PostgreSQL or MySQL database is wrong, but they have no idea how to find the cause given the complexity (and time limitations from the myriad of other things they have to focus on). The health checks can proactively identify problems such as missing indexes or cache misses, among others. Table-level health checks — like data bloat — provide even more fine-grained, automatic checks to help users avoid resiliency and performance issues. Identifying these problems through health checks is just the first step — OtterTune is working on additional features that will enable one-click remediation of surfaced issues.</p><p>As cloud database adoption continues to skyrocket, we believe OtterTune will be the key optimization platform upon which databases run efficiently, cost-effective and with high degrees of performance and resilience. Automatically delivering this as a platform requires a deep understanding of how applications use databases and database internals — which Andy, Dana, and Bohan have in spades. We’re thrilled to <a href="https://techcrunch.com/2022/05/10/2309852/">co-lead OtterTune’s Series A</a> alongside our friends at Race Capital, together with Accel participating.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c8f7b5bd22e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[It’s 10 PM, Do You Know How Reliable Your Systems Are?]]></title>
            <link>https://medium.com/@NickWashburn/its-10-pm-do-you-know-how-reliable-your-systems-are-a194cc983253?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/a194cc983253</guid>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[chaos-engineering]]></category>
            <category><![CDATA[continuous-verification]]></category>
            <category><![CDATA[site-reliability]]></category>
            <category><![CDATA[distributed-systems]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Thu, 03 Feb 2022 23:59:45 GMT</pubDate>
            <atom:updated>2022-02-03T23:59:45.210Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9DR6daWdJIxQBZ1kyy8AOA.png" /></figure><p><strong>By Nick Washburn &amp; Andy Fligel</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/its-10-pm-do-you-know-how-reliable-your-systems-are/"><em>Intel Capital website</em></a><em> on January 18, 2022]</em></p><h4><strong>Our Investment in Verica, the Continuous Verification Company</strong></h4><p>In the Fall of 2019, we were reading <a href="https://www.oreilly.com/library/view/chaos-engineering/9781491988459/">O’Reilly Chaos Engineering</a>, the then-current book about the principles of Chaos Engineering first developed by Casey Rosenthal and his team at Netflix. As we digested this introductory book, it became abundantly clear to us that we needed to meet him. Casey previously ran the Traffic &amp; Chaos Team at Netflix, helping to develop <a href="https://netflixtechblog.com/chap-chaos-automation-platform-53e6d528371f">the Chaos Automation Platform (ChAP)</a>. Beyond that, he and his team were ultimately responsible for keeping Netflix “up”, effectively moving more bytes across the internet than any other team on the planet (not to mention across a <a href="https://aws.amazon.com/solutions/case-studies/netflix/">massive amount of AWS instances</a>).</p><h4><strong>Chaos Engineering Can’t Be Limited to Netflix</strong></h4><p>We were full of questions, but one came front and center: Why was Netflix the only company practicing chaos engineering at scale? It seemed counterintuitive to us considering it was solving an obvious pain point that every company endures, regardless of size.</p><p>Every company values reliability, software and services should be available and dependable. When systems go down, for whatever reason, business is impacted and it often results in a fire drill of epic proportions -&gt; customer complaints, SRE beepers buzzing, post-mortem incident reports and analyses, and band-aid fixes until the next incident comes around. And this is all with the backdrop that the complexity of those very systems continues to increase. The reality is incidents are costly and pervasive.</p><p>Yet it seemed from afar that only Netflix was taking a truly proactive approach to try and experiment with their software and systems — in production — to test hypotheses of how those systems would react to real world failures. Why were most companies sticking to reactive tactics (i.e., detection, remediation, disaster recovery), when a proactive approach was needed to identify vulnerabilities in increasingly complex systems before they occur?</p><h4><strong>A Bit of Schooling from the Experts</strong></h4><p>After we were introduced to Casey, he brought along his good friend and co-founder at Verica, Aaron Rinehart. Aaron was previously the Chief Security Architect at UnitedHealth Group and led the DevOps and open-source transformation within the company as well. Taking this multi-disciplinary expertise, Aaron also pioneered the development of security chaos engineering. Not only does every company value reliable services that are always available, but they need those services to be secure as well!</p><p>As we met with Casey and Aaron, it was clear they were at the forefront of pushing the industry forward in these disciplines. Rather than simply speaking about chaos engineering, they quickly focused our minds on the concept of Continuous Verification, an experimentation platform that allows teams to proactively and safely discover system weaknesses (whether availability or security) before they disrupt business outcomes. Given we spent a lot of time in DevOps at Intel Capital, they really summarized this nascent, but highly-compelling discipline to us by tying into the natural evolution of the CI/CD process:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Qb2ETa-tWIfBxwxR1zdRvg.png" /><figcaption>Source: Verica</figcaption></figure><h4><strong>It’s Not Just Breaking Things in Production</strong></h4><p>When asked why more companies are not doing this, Casey and Aaron provided a compelling narrative of the challenges for enterprises in tooling and instrumenting principles of chaos engineering and continuous verification in a safe way. It’s hard! But what struck us was not merely the technical difficulties of a DIY approach, but the organizational headwinds that prevent a lot of companies from undertaking this effort. Some are driven by fear or misunderstanding (“I don’t want to break things in production”). Some are time (“I don’t have time to break things in production”).</p><p>Most importantly, however, Casey and Aaron made it very clear it’s <strong><em>not</em></strong> about just breaking things in production. You don’t go kick your TV as hard as you can over and over to see if it would break — beyond the stupidity of that, the reality is you would learn nothing useful (well, aside from how hard you can kick a TV). Casey and Aaron explained the entire point of chaos engineering and continuous verification is not to break things — it is to understand things. By using <strong><em>empirical experimentation</em></strong> to proactively discover vulnerabilities in software and systems, teams can better understand how their systems react to real world conditions. SRE, DevOps and DevSecOps teams can start with a hypothesis of how their system should act, test that hypothesis, and modify their system based upon the results.</p><p>Beyond the philosophy of what continuous verification is intended to do, Casey and Aaron also deeply understand the features that enterprises need to safely deploy and operationalize such a platform. The Verica Continuous Verification Platform is an enterprise-first platform with the right abstractions and useability to help companies instrument such principles in a safe and controlled way.</p><h4><strong>Our Continued Investment in Continuous Verification</strong></h4><p>Since first investing in Verica’s seed round and subsequently leading the <a href="https://venturebeat.com/2022/01/18/continuous-verification-company-verica-raises-12m-to-make-systems-more-resilient/">Series A round</a>, we’ve been thrilled to watch the progress the Verica team has taken with their vision. Large enterprise customers are relying on Verica’s platform to empower their teams to gain confidence in their software and systems. The platform has an enterprise style deployment, with Kubernetes and Kafka integrations. The feature roadmap is compelling with exciting announcements in the near term.</p><p>Beyond the commercial progress, the team has continued to push the industry forward, including Casey working with Nora Jones on an updated <a href="https://www.oreilly.com/library/view/chaos-engineering/9781492043850/">O’Reilly Chaos Engineering</a> book, Aaron working with Kelly Shortridge on the first <a href="https://www.oreilly.com/library/view/security-chaos-engineering/9781492080350/">O’Reilly Security Chaos Engineering</a> book, and Courtney Nash, James Wickett and the broader Verica team on launching <a href="https://www.thevoid.community/">Verica Open Incident Database</a>, a community-contributed collection of software-related incident reports to facilitate discussion and learnings from past failures. Lastly, the <a href="https://chaos.community/">Chaos Community Broadcast</a> of course!</p><p>We’re thrilled to take the covers off our work with the Verica team and excited for things to come!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a194cc983253" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Anyscale: The Infinite Laptop for AI]]></title>
            <link>https://medium.com/@NickWashburn/anyscale-the-infinite-laptop-for-ai-bb0d36ab6673?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/bb0d36ab6673</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[distributed-systems]]></category>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[open-source-software]]></category>
            <category><![CDATA[distributed-compute]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Thu, 03 Feb 2022 21:58:05 GMT</pubDate>
            <atom:updated>2022-02-03T21:58:05.303Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DI-vcz8Es0QUvHGETkzPEA.jpeg" /><figcaption>Anyscale Co-Founders: Robert Nishihara, Ion Stoica &amp; Philipp Moritz (left to right)</figcaption></figure><p><strong>By Assaf Araki &amp; Nick Washburn</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/anyscale-the-infinite-laptop-for-ai/"><em>Intel Capital website</em></a><em> on December 8, 2021]</em></p><p>Enterprises investing time and money into artificial intelligence (AI) and machine learning (ML) is not about buzzwords or building out cool tech — it’s about finding new business value. IDG’s <a href="https://www.idg.com/tools-for-marketers/2021-state-of-the-cio/"><em>2021 State of the CIO</em></a> report states that CIOs are tasked with creating new revenue streams and uplifting existing ones. It’s what public shareholders expect, and is a main topic of discussion in boardroom conversations. Management teams in turn look to a variety of teams including data scientists, data engineers and platform engineers, in addition to all the other teams that touch the AI/ML lifecycle, to implement new AI/ML practices and expect quick results. While AI/ML continues to grow at various points on the <a href="https://www.gartner.com/en/newsroom/press-releases/2021-09-07-gartner-identifies-four-trends-driving-near-term-artificial-intelligence-innovation">Hype Cycle</a>, there continues to be a disconnect between the business results <strong>expected</strong> from operationalizing AI/ML, and the tooling and steps required to <strong>actually</strong> operationalize AI/ML.</p><p>Meanwhile, there are some deeply technical realities that are happening behind the scenes from those management teams:</p><ul><li>Data, and especially training data, is the oil that fuels the AI industry. Unlike oil in the physical world, the amount of data is growing exponentially. So is the complexity of ML algorithms. The pace of innovation is both exciting and suffocating in some cases.</li><li>More data and complex algorithms require larger machines with more compute and memory. Per <a href="https://openai.com/blog/ai-and-compute/">research</a> conducted by OpenAI in 2018, deep learning compute demand grew 35x every 18 months from 2012–18, and memory demand grew 20x every 18 months from 2016–18 (the release of GPT further increased this trend).</li><li>The AI/ML pipeline continues to encompass many heterogeneous workloads. Exploration (training) and inference (production) phases have different characteristics. For instance, the exploration phase is batch-oriented, using huge amounts of data, while the inference phase is real-time on small amounts of data (e.g., one record or one image). The exploration phase is further split into pre-processing and ML training that each use different libraries and may use different distributed systems.</li></ul><p>It’s the perfect storm: executive-driven business needs with difficult timelines, AI-centric business applications, increasing complexity and heterogeneity, technical limitations, and inadequate tooling.</p><p>What developers need is a cloud “easy button”, where an individual developer can write code once on their laptop and scale it to the cloud seamlessly without any refactoring or modifications. A serverless experience, without the limitations imposed by cloud service providers (CSPs) at run-time.</p><h4>The Solution: Ray</h4><p>At Intel, we have long collaborated with academic centers of excellence across the globe, including UC Berkeley. This collaboration extends from <a href="https://www.intel.com/content/www/us/en/research/overview.html">Intel Labs</a> to various business unit engineers. We seek not only to conduct state of the art research, but also to apply that research to real-world problems. Of course, at Intel Capital we work closely with our partners in Intel Labs and broader Intel, tracking the research projects and corresponding open source communities from the earliest of days.</p><p>As sponsors of UC Berkeley <a href="https://amplab.cs.berkeley.edu/">AMPLab</a> and early contributors to the BDAS stack, we had a deep understanding of the benefits of the new data analytics stack, but we also understood its limitations.</p><p>Five years ago, towards the end of AMPLab and the birth of UC Berkeley’s <a href="https://rise.cs.berkeley.edu/">RISELab</a>, Assaf Araki had a meeting with Robert Nishihara and Philipp Moritz, then PhD students at UC Berkeley working under Professor Ion Stoica on a new open-source project called <a href="https://www.ray.io/">Ray</a>. In its infancy, Ray was an experimental distributed extension of Python that made it easy to write ML applications that run on a cluster while providing the development and debugging experience of working on a single machine. Rapidly, Ray became the skeleton of the RISELab stack. Ray developed into a universal computing substrate to scale the most compute-intensive ML-workloads with the simplicity of a laptop-like development experience, all while leveraging a broad set of libraries and integrations that are built on Ray’s distributed execution framework. In fact, Intel’s IT team was one of the early adopters of Ray, leveraging it to scale Python, hyperparameter selection techniques, and auto modeling processes workloads with minimal code modifications. The Intel Artificial Intelligence and Analytics team further offered Ray as part of <a href="https://github.com/intel-analytics/analytics-zoo">AnalyticsZoo</a> and <a href="https://github.com/intel-analytics/BigDL">BigDL</a> and contributed to the open-source project by creating <a href="https://github.com/intel-analytics/analytics-zoo#getting-started-with-rayonspark">RayOnSpark</a> and <a href="https://github.com/oap-project/raydp">SparkOnRay</a>.</p><p>Here at Intel Capital, Ray reminds us of “Intel Inside”, a key interoperability point between hardware and applications. Ray offers one unified — and simple — platform to scale all of your AI and analytics libraries. With a robust and growing ecosystem of native and third party libraries, developers can use Ray as a single distributed platform for all of the heterogeneous tasks we outlined above: data ingestion, hyperparameter tuning, model training, model serving and so on. It reduces complexity, accelerates the time to deployment, increases the model accuracy, and shortens the time to retrain.</p><p>The growth of the Ray open-source community has also been incredible, speaking to the power of the platform. With more than <a href="https://github.com/ray-project/ray">18,000 GitHub Stars</a> for Ray core and contributor growth continuing to rise rapidly, we view Ray’s adoption as best-in-class.</p><h4>Anyscale: Our Partnership Since Formation</h4><p>As we monitored the development of Ray and the associated libraries within RISELab, we quickly developed the thesis that this is the solution to the perfect storm of operationalization of ML complexities. The path to success is distribution, and the path to simplicity in distribution is Ray and Anyscale.</p><p>Both Assaf and Nick Washburn recall driving over to UC Berkeley in early 2019 to meet with Professor Stoica to discuss the incorporation of <a href="https://www.anyscale.com/">Anyscale</a> and how Intel Capital would love to start our partnership from Day 0, building upon the work that Intel Labs and the Intel technologists had already been working on with Robert, Philipp, Ion and the team for years. We were thrilled to participate in Anyscale’s <a href="https://techcrunch.com/2019/12/17/anyscale-ray-project-distributed-computing-a16z/">Series A round</a> in 2019 and <a href="https://techcrunch.com/2020/10/21/anyscale-adds-40m-to-bring-its-ray-based-distributed-computing-tech-to-the-enterprise-masses/">Series B round</a> in 2020, and we are equally excited to continue our strong support in today’s announcement of a <a href="https://www.forbes.com/sites/kenrickcai/2021/12/07/berkeley-research-lab-mints-second-billion-dollar-startup-anyscale/?sh=11508028866a">$100M Series C round</a>.</p><p>Since our initial investment in 2019, we’ve loved working with the Anyscale team and watching the growth in not only the community around Ray and its associated libraries and integrations, but also the development of Anyscale’s <a href="https://www.anyscale.com/product">managed offering</a>, which was also just announced in general availability. Anyscale’s product truly creates the infinite laptop experience for a developer. Developers can write a ML application on their laptop and — with zero code changes or refactoring — scale across the cloud in a serverless experience without run-time limitations. It offers multi-cloud portability, offering users the ability to click a button to move workloads on any Anyscale managed cloud offering. It is the easy “on ramp” to the superhighway of ML workloads, offering simple APIs for job automation and dependency and cluster management.</p><p>More exciting work from the Anyscale team to come. Onward!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bb0d36ab6673" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Empowering Developers with a Universal Cloud API: Our Investment in Upbound]]></title>
            <link>https://medium.com/@NickWashburn/empowering-developers-with-a-universal-cloud-api-our-investment-in-upbound-d7a7a62acf83?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/d7a7a62acf83</guid>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[control-plane]]></category>
            <category><![CDATA[cloud-native]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[open-source-software]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Thu, 03 Feb 2022 21:46:58 GMT</pubDate>
            <atom:updated>2022-02-03T21:46:58.177Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="Bassam Tabbara, Founder &amp; CEO of Upbound" src="https://cdn-images-1.medium.com/max/1024/1*JLPJ8DueTQa471syZAs03g.png" /><figcaption>Bassam Tabbara, Founder &amp; CEO of Upbound</figcaption></figure><p><strong>By Nick Washburn &amp; Andy Fligel</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/empowering-developers-with-a-universal-cloud-api-our-investment-in-upbound/"><em>Intel Capital website</em></a><em> on November 30, 2021]</em></p><p>Within enterprises, the goals across the software delivery lifecycle are fairly straightforward. Business leaders want to increase revenue and customer retention. Application developers want to ship business logic in an agile fashion. Lastly, DevOps or infrastructure engineers want to provide application developers with a “self-service” experience that increases their agility and ability to deliver value.</p><p>In practice, however, achieving those goals is rather difficult. The entire DevOps movement, for example, arose from the need to break down the barriers between application developers and infrastructure teams to speed up deployments. While strides were made in automation, it often came at the cost of complexity, which actually slowed things back down.</p><p>One could even argue the same thing about the public cloud. In March 2006, Amazon launched AWS and <a href="https://press.aboutamazon.com/news-releases/news-release-details/amazon-web-services-launches-amazon-s3-simple-storage-service#:~:text=SEATTLE%2D%2D(BUSINESS%20WIRE)%2D%2D,.amazon.com%2Fs3">S3 object storage</a>, kicking off the public cloud “IaaS” area. Since then, all major cloud service providers (CSPs) have vertically integrated, offering a host of solutions from underlying infrastructure (i.e., compute, networking), to services (i.e., databases, developer tools), to entire platforms (i.e., E2E machine learning platforms). Today, AWS has <a href="https://aws.amazon.com/products/?aws-products-all.sort-by=item.additionalFields.productNameLowercase&amp;aws-products-all.sort-order=asc&amp;awsf.Free%20Tier=*all&amp;awsf.tech-category=*all">222</a> separate products. In parallel, a host of independent software vendors (ISVs) have built their own cloud-based services (i.e., data warehouses, CDNs) that can run on any public or private cloud infrastructure. For example, Snowflake’s Cloud Data Platform runs completely on cloud infrastructure and can be <a href="https://docs.snowflake.com/en/user-guide/intro-cloud-platforms.html">hosted on AWS, GCP or Azure</a>.</p><p>The result? A highly fragmented cloud ecosystem with a myriad of offerings at various levels of abstraction. Enterprises, in turn, have increasingly heterogeneous IT estates (i.e., hybrid, multi-cloud) with many vendors. And with the added complexity of a rise in regulation, the promise of “self-service” for application developers appears as a fleeting myth.</p><h4>The Control Plane: The Key to the Cloud</h4><p>The vertical integration driven by the CSPs has largely been enabled by one thing: the control plane. Each CSP has its own proprietary control plane upon which its respective platform is built. That control plane enables the CSP to expose a uniform application programming interface (API) and common necessities when running large service catalogues to end-users; think access control, billing and metering, governance, etc.</p><p>When Google open sourced <a href="https://kubernetes.io/">Kubernetes</a> in 2014, it promised to automate the deployment and management of cloud workloads running in containers. Today, Kubernetes is the de facto container orchestration technology of choice, but there is still a lot of complexity in deploying and managing cloud workloads.</p><p>However, Bassam Tabbara, Founder &amp; CEO of <a href="https://www.upbound.io/">Upbound</a>, who rightfully identified control planes as the best kept secret in the cloud, realized that the power of Kubernetes was in its control plane and could be extended well beyond container orchestration. It’s because of deeply technical and forward thinking visionaries like Bassam, who helped standup Azure at Microsoft, that cloud innovation continues at the pace it does.</p><p>In 2018, Bassam and the team at Upbound created <a href="https://crossplane.io/">Crossplane</a>, an open source project that leverages Kubernetes to offer users a more compelling value proposition, namely a universal cloud control plane that automates away a lot of complexity. Through a single, declarative API, Crossplane users can provision cloud infrastructure (i.e., VMs, containers) as well as cloud services (i.e., databases) across hybrid cloud estates. And unlike legacy infrastructure-as-a-code tooling, Crossplane, once provisioned, continues to monitor resources and automatically reprovisions resources if it detects any drift from the desired state.</p><h4>A New DevOps Standard</h4><p>We expect broad adoption of multi-cloud control planes and believe Crossplane will be a winner in this category, helping platform teams finally achieve the original goals of DevOps to provide a “self-service” experience to application developers. With Crossplane, platform teams and developers share a common API across cloud environments and developers enjoy the speed and ease of self-service access to the broad array of cloud resources they need with the guardrails set by the platform teams.</p><p>Donated to the Cloud Native Computing Foundation (CNCF) in 2020, Crossplane recently graduated from <a href="https://blog.crossplane.io/crossplane-cncf-incubation/">sandbox to incubating project</a>, reflecting rapid industry adoption and support from the likes of AWS, Microsoft Azure, GCP, Alibaba, and IBM Red Hat among others. As the creators of Crossplane and commercial offerings around it, Upbound is uniquely positioned to build on this momentum and help customers achieve the most value from the technology, whether deployed on-prem or consumed through SaaS.</p><h4>A Visionary CEO</h4><p>Today, Upbound <a href="https://techcrunch.com/2021/11/29/upbound-grabs-60m-series-b-to-grow-open-source-crossplane-cloud-management-project/">announced</a> $60M of new funding, including the Series B round which we are thrilled to have participated in. Beyond the technology and market opportunity, we are very excited to partner with Bassam. We’ve had a chance to get to know Bassam over several years, and his vision for Upbound has always impressed us. Upbound launched its commercial offerings in Q2 with strong customer reception, an early validation of the opportunity Bassam foresaw. We look forward to helping Bassam fully realize the ambitious vision he has laid out.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d7a7a62acf83" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Much Needed “Easy Button” for Modern Network Operations: Our Investment In Augtera Networks]]></title>
            <link>https://medium.com/@NickWashburn/the-much-needed-easy-button-for-modern-network-operations-our-investment-in-augtera-networks-ac31dd2eb57c?source=rss-481411f3020c------2</link>
            <guid isPermaLink="false">https://medium.com/p/ac31dd2eb57c</guid>
            <category><![CDATA[intel-capital]]></category>
            <category><![CDATA[cloud-network]]></category>
            <category><![CDATA[network-management]]></category>
            <category><![CDATA[netops]]></category>
            <category><![CDATA[aiops]]></category>
            <dc:creator><![CDATA[Nick Washburn]]></dc:creator>
            <pubDate>Thu, 03 Feb 2022 21:34:29 GMT</pubDate>
            <atom:updated>2022-02-03T21:34:29.524Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ib0_tZqzzWFrfR3RhRCW0A.jpeg" /></figure><p><strong>By Nick Washburn &amp; Andy Fligel</strong></p><p><em>[This post originally appeared on the </em><a href="https://www.intelcapital.com/the-much-needed-easy-button-for-modern-network-operations-our-investment-in-augtera-networks/"><em>Intel Capital website</em></a><em> on June 29, 2021]</em></p><p>Let’s get real about the current state of network operations. It’s a total mess. It’s 1,000s of daily alerts and alarms based on static thresholds that are largely set by trial and error as a result of some problem that occurred in the past. Remember those financial product disclaimers, “past performance is not indicative of future success”… those should be applied to network operations. And worst of all, it’s the beeper buzzing at 2 A.M. for a poor SRE or NetOps team member that needs to come in and spend hours debugging a grey failure (i.e. network performance degradation) all while the application being served on top is driving a poor experience for end-users. This leads to unhappy customers, lost business value, endless fire drills, and occasionally lost jobs. Fundamentally, it’s all reactionary; the problem has already occurred and we need to figure out how to stop it ASAP before it gets worse.</p><p>Now let’s click down a bit more precisely on how we got here, before we talk about what we’re going to do about it. There are a number of critical trends that we’ve been observing the past couple of years:</p><ul><li><strong>Enterprise Networks are Heterogeneous &amp; Large</strong>. The complexity is beyond “human scale” to comprehend. The modern enterprise network is made up of many vendor products, both virtualized and appliance based. There are an untold amount of devices running in production, each with different model and firmware combos. Layer on top the unique and independent operating systems that are running in these environments (some vendor proprietary, some open source). A typical Fortune 500 datacenter network is running in the 1,000s to 10,000s+ of appliance-based and virtualized switches, routers, firewalls, and load balancers, not to mention the corresponding servers. These devices, in turn, have 2 to 3 billion lines of text in configuration coding with 100s to 1,000s of rules per device. On top of that, the NetOps team is running 100s of changes per week in this environment. And the scenario above is talking about a single datacenter. The introduction of new, dynamic and cloud-native network architectures resulting from microservices and containers, hybrid networking, and cloud to edge migration further exacerbates the issue. Fundamentally, network operators within large enterprises and service providers have a gargantuan hybrid network stack that is utterly complex.</li><li><strong>They are Managed with Inadequate Tooling</strong>. To put it bluntly, management is extremely difficult, highly manual and largely reactionary once problems occur. One culprit is the tooling being used. Many of the existing Network Performance Management &amp; Diagnostic (NPMD) tools still in production with many NetOps teams were not originally built to handle the scale and complexity of the networks they run today. Meanwhile, telemetry collection often provided by these legacy NPMD tools is largely insufficient.</li><li><strong>This Results in Fragility</strong>.<strong> </strong>When management is extremely difficult, highly manual and largely reactionary… well… problems occur. Identifying problems is hard, and then fixing them without then creating yet another problem is harder. But fragility is not just failure, it is also confidence. How can you have confidence in the performance of your systems when you are receiving 100s to 1,000s of alerts a day, the vast bulk of which are not relevant or actionable?</li></ul><blockquote><strong>To sum it up, scale + complexity + many changes + bad tooling = degradations, outages and, ultimately, unhappy customers.</strong></blockquote><p>With this backdrop in mind, we’ve been on the hunt for the right team and product to really take a holistic approach to entirely changing the paradigm of modern network operations. Enter <a href="https://augtera.com/">Augtera Networks</a>, which we’re thrilled is launching out of stealth today with a <a href="https://venturebeat.com/2021/06/29/augtera-networks-raises-13m-to-apply-ai-to-networking/">$13M Series A round</a> led by Intel Capital with participation from Bain Capital Ventures, Dell Technologies Capital and Acrew Capital.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pjbZ_Ny-0HgSUjWvfpgPWg.jpeg" /><figcaption><em>Rahul Aggarwal, Founder and CEO, (Left) and Bhupesh Kothari, Co-Founder and VP of Engineering (Right)</em></figcaption></figure><p>When we first met Rahul Aggarwal in the Fall of 2019, it was an immediate meeting of the minds. Rahul walked through his views on the current challenges of network operations today, which matched our observations entirely. As Rahul described, the existing tooling was just collecting signals from cloud, WAN and datacenter networks, and placing them in a massive (and ever growing) data “haystack”. Meanwhile, network operators were manually plowing through this “haystack” to find the “needles” of data that really can remediate a problem.</p><p>Fortunately, Rahul and his co-founder, Bhupesh Kothari, have a deep understanding of network architecture to re-think the approach to operations. Having first met at Juniper, Rahul and Bhupesh have been instrumental in developing numerous modern networking protocols and methodologies. At Augtera, they built a Network AI Platform from the ground up, bringing the benefits of AI-driven operations, planning and orchestration to physical, virtual and cloud network environments.</p><p>Data is the core of the platform, ingesting standard and vendor specific data at large scale. In parallel, unsupervised machine learning models bring automated and proactive management, identification and remediation of grey failures and other degradations before they impact customers. The signal to noise ratio is dramatically increased, with alerts being highly actionable. Some of the use cases that are, or will be in the near-future, covered by Augtera’s Network AI Platform include:</p><ul><li><strong>Automated Network Topology</strong> — Providing network operators with zero touch, multi-layer topology with deep predictive visibility</li><li><strong>Automated Grey Failure Detection</strong> — Early and predictive detection of grey failures in network hardware and the control plane</li><li><strong>Automated Multi-Layer Correlation</strong> — Correlation of network events and Augtera-generated AI insights to reduce root cause analysis times by an order of magnitude and more</li><li><strong>Recommended Remediation</strong> — The platform will soon be extended to learn on an unsupervised basis from past fixes and automatically recommend fixes to new network incidents</li><li><strong>Automated Remediation</strong> — The platform currently has direct integrations with ticketing systems, and will also soon be extended to allow operators to apply recommended fixes automatically.</li></ul><p>Augtera’s Network AI Platform is driving towards offering customers an “easy button” for managing networks in a proactive and automated way. Rahul and Bhupesh have leveraged their deep domain experience to build a product that is automatically identifying and correlating grey failure and other challenging networking problems that have plagued large enterprises and service providers. Soon the platform will be extended to provide recommended and automated remediation of these challenges as well.</p><p>The feedback from our customer calls on the ROI they received from the platform really validated our thesis that the market was in need of a wholesale new approach — which Augtera is delivering. Augtera is currently deployed in large-scale production environments with several major enterprise and service provider customers, including both hybrid datacenter, WAN and backbone networks. Each of the customers was originally focused on detecting and fixing grey failures and other hidden network misbehaviors before they caused customer-facing issues and/or outages. As we clicked-down on these deployments, it was clear that Augtera’s Network AI Platform was central to their networking strategy and provided significant business value, including identifying numerous network anomalies that would have impacted service had they not been proactively detected. Another customer noted that they were seeing between 25% to 75% daily reduction in tickets compared to traditional NPMD tools. Most importantly, all of these systems are running at large scale, and Augtera’s unsupervised machine learning platform is continuing to learn from these networks and gain further efficacy.</p><p>We’re thrilled to have established our partnership with the Augtera team at the seed stage and now lean in to lead the Series A. Together with our co-investors, we’re looking forward to the new future for network operations moving from manual and reactionary to automated and proactive with Augtera.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ac31dd2eb57c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>