Obsessed with developer tooling and steep trails.

Ever since I was young, I've loved to explore, learn and build, mostly with tech. At 12 I wrote my first SaaS app, and to my surprise it brought in several hundred dollars a month through Google AdSense.

The only thing I loved more than computers as a kid was sports and the outdoors. That took me to university in Hawaii to play NCAA soccer. Almost 20 years later not much has changed, it just got steeper. These days it is skyrunning: long days on trails that go up far more than they go along, with Mo, my Kangal, setting the pace. It is still the one thing that reliably pulls me away from my desk.

I am not really a runner, though. I have been an athlete most of my life, but running on its own never brought me much joy, and I have never done it professionally. I know I will never be the fastest one on the mountain. I also know I will not quit. The point was never the running. It is showing that a strong mind with a clear goal can get you almost anywhere. It is also good for the dog.

Next up is the Pacific Crest Trail, northbound: 2,650 miles at an average of 60 a day. I do not know yet if that pace holds, and finding out is half the reason to go. After that, maybe a race or two, something like the Tor des Géants, though 2,650 miles will probably change what I want out of all this. I will share all of it here, including the days that do not go to plan.

For the last 15+ years I have built personal and private projects, from small apps to large enterprise services, which is where my obsession with scalable systems comes from. I went deep on data storage and transfer, and on distributed and decentralized compute. That is a big part of why I started Stacks.

The goal behind all of it is software that is as fast and energy efficient as it can be. In the age of AI that matters more than ever: every request an agent makes runs on real hardware and real power, and most of it passes through ordinary code on the way. Good abstractions are one of the few levers that scale. Get the defaults right once, in something like Stacks, and every app built on top does less work for the same result.

What has never changed is where the work comes from. Every project I have started began as something I, my team or the community around us actually needed, where only paid tools did it well and nothing open source came close. Stacks, because I kept rewriting the same infrastructure under every app. The athlete software, because I wanted to use it myself. That need gives each project a clear reason to exist, and it is what keeps me working on it.

The two halves feed each other more than you would think. A long day in the mountains is mostly pacing, small decisions compounding, and being honest with yourself about what is actually working. That is also a fair description of shipping software. So is refusing to quit, especially when it comes to Open Source.

Today that means Stacks and its broader ecosystem, and a growing amount of agentic tooling. The tools engineers use are quickly becoming the tools their agents use, and that changes what a good one looks like. On top of it sit the apps, lately HQ.training and Wildloop: athlete software I wrote because I needed it to exist. Follow along on GitHub for the latest.