I am a software engineer working at the seam between large language models and the systems they run on. Most of what I write here comes out of that work: agent harnesses, NixOS configuration, TypeScript, and whatever broke that week.

I have no formal training in computer science. I learned by breaking installs, reading source, and rebuilding the same configuration until it held. That path makes you suspicious of anything you cannot reproduce from scratch, which is why this site is more about reproducible systems than about frameworks.

Philosophy meets practice

I believe that large language models represent a fundamental shift in how we approach problem-solving and knowledge synthesis. My work focuses on bridging the gap between theoretical AI capabilities and practical software engineering solutions, always with an eye toward systems that are as intellectually satisfying as they are functionally robust.

In the trenches

My autodidactic journey has taken me through the depths of functional programming paradigms, the elegance of immutable infrastructure, and the fascinating complexity of emergent AI behaviors. I approach each project as both an engineering challenge and an opportunity to push the boundaries of what's possible when human insight meets machine capability.

Despite being labeled an LLM advocate, I remain skeptical of the hype—drawing from the doctrine of three bodies in Indian philosophy, I believe true intelligence requires the integration of physical, subtle, and causal embodiments that current AI fundamentally lacks.

While I'm passionate about Linux and NixOS, I believe everyone should use whatever tools work best for them—my enthusiasm stems from personal preference, not evangelism.

This site is built with

Performing reactivity and statefulness, special guest SolidJS