Writing
From the lab
Practical notes on browser-native AI. Opinionated, technical, no fluff.
WebGPU: the browser's new compute primitive
How WebGPU changes what's possible for AI in the browser, and why it matters for local inference at scale.
Building browser RAG with Transformers.js
A practical walkthrough of retrieval-augmented generation running entirely client-side — embeddings, chunking, and retrieval without a server.
On-device embeddings: a practical guide
Generating, storing, and querying text embeddings without leaving the browser tab. Includes a deep dive on IndexedDB as a vector store.