←Back
/dev/michael
michaellivs.com41 posts5 readersStale
Thoughts on building with AI agents, Claude Code, infrastructure, and the craft of software engineering. By Michael Livshits.
* Latest posts *
- [JAN.24.2026]Playing Capture the Flag with Claude↗
- [JAN.23.2026]Claude Code Tasks: One Less Dependency↗
- [JAN.23.2026]llms.txt Doesn't Do What You Think↗
- [JAN.21.2026]I Understand My Code. I Just Don't Know It.↗
- [JAN.19.2026]Why I Chose FTS Over Vector Search for Claude Code Memory↗
- [JAN.18.2026]Open Responses Solves the Wrong Problem↗
- [JAN.16.2026]Claude Quest: pixel-art visualization for Claude Code sessions↗
- [JAN.14.2026]Skills aren't the innovation↗
- [JAN.12.2026]psst v0.2.0: Environments, Scanning, and psst run↗
- [JAN.11.2026]Why Anthropic and Vercel chose different sandboxes↗
- [JAN.10.2026]Claude forgot. ran didn't.↗
- [DEC.26.2025]psst v0.1.3: Now It Actually Masks Your Secrets↗
- [DEC.23.2025]Split Personality Claude↗
- [DEC.22.2025]psst 🤫 Because Your Agent Doesn't Need to Know Your Secrets↗
- [DEC.20.2025]The Agent Harness↗
- [DEC.19.2025]Context Engineering Has No Engine↗
- [DEC.15.2025]Tool Design is All About the Flow↗
- [NOV.29.2025]Reverse-engineering Claude's sandbox, then building my own↗
- [NOV.28.2025]Embedding Claude Code sessions in blog posts↗
- [NOV.28.2025]Anatomy of agentic systems↗