Archive

Every published note, link, TIL, and post, in reverse chronological order.

2026

  1. My Ray Summit Talk post
  2. Apparently 32 PRs a Week Was Conservative post
  3. My Day Job Is Unsupervising Agents post
  4. A Frustrating AI Code Smell: Every Problem Decomposes to a dataclass With Four Reasons post
  5. Completely Impractical but Delightful: An E-Reader Attached to My Phone post
  6. If AI Coding Trends Continue, I’ll Be Merging 20 PRs a Day post
  7. I Put My Dog in My Coding Agent post
  8. Dogfooding Is the New Code Review link
  9. An Update on Codex's Gremlins post
  10. There are Gremlins lurking in Codex GPT-5.5 link
  11. The Shiny Object Trap post
  12. Karpathy's LLM wiki looked familiar. Here's how to bootstrap your own in 5 minutes. post
  13. Rummaging through Claude Code's spring cleaning closet post
  14. It shouldn't take 50,000 buttons to format a bulleted list note
  15. It takes 7,500 tests, 12 languages, and 3 browsers to make text wrap around a circle post
  16. Agent-Friendly CLI Design link
  17. LLMs Are Cowboy Coding link
  18. Don't RAG Until You Have To note
  19. ML Monitoring: Keep It Simple, Keep It Compounding post
  20. Ruining the Fun of March Madness by Having AI Fill Out My Bracket post
  21. Brandolini's Law Is Flipping note
  22. Don't Replace Your Job, Replace Your Chores post
  23. "The CLI is Dead" note
  24. My AI Dev Flow in 10 Bullet Points note
  25. Agents Only Need Four Tools note
  26. Notes on Self-Hosting LLMs (Early 2026) post

2025

  1. Muon: The Magic of Three Numbers Making GPUs Go Brrr link
  2. How Claude Code actually uses CLAUDE.md files link
  3. Homelab Update: The Paternity Leave Edition post
  4. Firecrawl: A Simple Web Scraping Script note