Blog

Why is matrix multiplication the bottleneck in LLM inference? We formulate the rate-distortion probl...

Introduction to Anthropic's Claude Cookbooks repository. Covers classification, summarization, and t...

A technical analysis of Palantir's core architecture. Why the Ontology is distinct from a Semantic L...

Devin 2.0 moved from $500/month to $20/month. Here's what that implies about product packaging, eval...

Why can we learn some functions with few samples but exponential time? The gap between information-t...