anurag builds things

anurag builds

I build things to understand how they actually work. A question, a system built to answer it, a comparison against real data. Just the code, the curves, and what they actually mean.

A painted still-life of a leather-bound notebook with logos for OpenAI, Anthropic, Microsoft, Mistral, GitHub, Cursor, Zed, and Google Gemini arranged across its cover, on a worn wooden desk with a brass magnifying glass, a coiled USB cable, a fountain pen, and an ivory card bearing a single lime-green dot.

The AI Tools I’ve Actually Used

ai-coding
living-document
timeline

A running vertical timeline of the AI tools I’ve leaned on day-to-day - for coding, chat, research, and the work in between. Updated as I adopt new ones.

22 May 2026

Prompts I Keep Coming Back To

prompts
claude-code
ai-coding
living-document

A commonplace book - prompts and magic words I’ve collected from others and keep reaching for. Not mine; just my favorites. Growing as I find more.

21 May 2026
YouTube thumbnail for the companion video: '21 agents, 1 human, prod K8s driver' over a tmux split-pane showing teammates working in parallel.

Claude Code Agent Teams on a Real Production Chore

claude-code
agent-teams
ai-coding
kubernetes
csi

I tried Claude Code’s experimental agent teams on an actual job - analyzing upstream divergence on Scality’s S3 CSI driver - not a toy demo. Field notes on what worked, what cost, and the three settings worth knowing.

03 Apr 2026
No matching items