Hi there! I’m Felix Arntz, a software engineer and open-source contributor focusing on the open web. I’ve been building things for the web for over 15 years, and I’m particularly passionate about AI and UX.
I am currently a Senior Software Engineer at Google, where I have been working on AI evals for web application code generation, and previously on various WordPress focused efforts, such as enhancing the CMS performance at scale, and launching the Site Kit project.
Additionally I am a core committer to the WordPress open-source project, and co-founder of both the WordPress AI Team and the WordPress Performance Team. I have spoken at over 20 WordCamps and other WordPress events.
A little history
I started building websites as a freelancer, which quickly evolved into more backend focused projects, building web applications with custom WordPress plugins at their core.
Since mid-2015, I have been heavily involved with contributing to the WordPress Core software. With WordPress version 4.6 I was recognized as a noteworthy contributor for the first time, and since November 2016 I have been a Core Committer.
Becoming more active in the broader WordPress community allowed me to collaborate with several companies in the ecosystem. This led me to eventually focus on contracting for these companies, in addition to maintaining my own client relationships. Some of the WordPress companies that I worked with during my freelance career are Yoast (I even have a Yoast avatar as you can see!), WP Engine, Aspen Grove Studios, and Usability Dynamics.
In 2018, I joined Google, from the very beginning in a CMS focused team primarily building WordPress solutions. For several years I was the lead engineer of the Site Kit plugin for WordPress, building the plugin from the ground up, leading it to launch, and growing it to over 4 million active installations.
In late 2021 I shifted my focus within Google, co-founding the WordPress Core Performance Team and subsequently being heavily involved in its projects to enhance performance of the WordPress project, to benefit all websites using the CMS.
In 2025 I started working on a Google project focused on evaluating and improving the web app code generation and coding assistance quality of Gemini. A key piece of that work is establishing a benchmark to assess web app code quality holistically.
As of May 2025, in a personal capacity, I am a co-founder of the WordPress AI Team and one of the co-Team Reps. With this team, the goal is to empower WordPress plugin developers to build AI features with ease, in a user-friendly and responsible manner.
What I enjoy when building software
I’m equally enthusiastic about advancing AI capabilities as I am about UX and DX. I love building intuitive APIs, and making the complex things simple. And when working on user-facing projects, making sure that their workflows can be done in as few steps as possible and provide fast and transparent feedback mechanisms. Whichever projects I’m working on or contributing to, my objective is to steer them towards these goals.
Implementing the components of a piece of software in a decoupled and reusable manner, and clearly scoping the software’s public API makes the entire software easier to maintain. This enables quick iterations with little risk of bugs. It even makes it possible to experiment without risking the integrity of your software.
Building software from decoupled components with distinct responsibilities also facilitates building performant software. By selectively loading only what’s needed you avoid overhead and keep the backend footprint low. From there, implementing interactive features following frontend performance best practices is almost a breeze, if you know what to pay attention to.
Last but not least, in terms of AI, my big passion (other than using it to accelerate software development workflows) is standardization. The AI landscape is moving so fast – which overall is a great thing – but this leads to lots of people and companies implementing 95% of the same thing in different incompatible ways. While this is a result of people wanting to build things fast (and thus minimize dependencies), it also leads to overall ecosystem progress being slower than it could be, because most innovative solutions have quite a bit of vendor lock-in, so it’s difficult for other solutions to build on top of it. Reinventing the wheel is the common result. We need to be cautious of premature abstraction – that can hurt progress just as much. But we need to identify the need for abstraction to accelerate progress. This is where projects like the AI SDK and the PHP AI Client shine.
Outside of software engineering
When I’m not coding, I am passionate about electronic music production, and I love going for long walks. You may also find me running around the world, or geeking out on movies, Mexican food (especially 🐟🌮), and beer. My guilty pleasure is Mountain Dew.
