Senior fullstack
engineer.
Senior fullstack engineer building AI-driven product systems.
Six years shipping production code — currently WellTheory, previously Digg and Sensible Weather. Based in Los Angeles.
// Now
What I'm into.
Updated monthly · last revised July 2026
- How to turn individual productivity gains from AI into cross-functional gains
- Balancing cognitive overload with cognitive engagement
- Which California wildflowers to plant in my backyard
// Featured
What I'm up to.
Led development of a modular content moderation pipeline designed to support iterative, pluggable models and capture distinct datasets for AI flagging, human review, and in-app moderation. Partnered with data engineering to build a RAG-based autonomous moderation layer, achieving 86.5% F1 (95% recall, 87% precision) and reducing the human review workload by 90%.
View all work ↗︎An interactive tool for exploring the inner mechanics of AI models. Currently features a live visualizer for transformer internals — attention heads, MLP layers, residual streams — using the mechanistic interpretability package nnsight and GPT-2 as a reference model. More experiences and components coming.
View portfolio ↗︎// About
A bit about me.
I'm a senior fullstack engineer based in Los Angeles, CA. I've spent the last six years shipping real production code — currently WellTheory, previously Digg and Sensible Weather.
I spend most of my energy right now on what good engineering actually looks like when an agent is doing the writing — what practices hold up, what paradigms break, and what changes when you're reviewing code you didn't write but are responsible for. A lot of what made me good at the job before is still useful, some of it is in the way, and most of the work right now is figuring out how to use the good parts in this new iteration of the job.
Outside of work, you can usually find me hiking, reading, listening to records or playing with my dog.