FRAN CANETE
02 / Notes · How I work with AI

How I build software with coding agents.

Not a blog. A small set of notes on the practices, tools, and patterns I lean on when I'm building with AI in the loop. The stuff I'd rather write down once than keep explaining. Slow-moving by design.

01 / my claude code setup Live

My Claude Code setup.

A static note on how I keep my coding-agent workflow version-controlled: prompts, commands, conventions, and the review loop that makes Claude Code useful day to day.

02 / my pi agent setup Live

My Pi agent setup.

A static note on how I am starting to use Pi as my coding-agent harness: the model defaults, subagents, memory, web tools, and the small workflow rules that make it useful.

03 / how i keep coding agents honest Live

How I keep coding agents honest.

A static note on my review loop for AI-generated code: using cold validators, scoped reviews, evidence, and fresh-context subagents before I trust the work.