~$ ls -la ~/projects
Demos, hands-on training, and self-built projects from my ongoing cloud studies.
~$ tail -f updates.log
~$ cd ai/ && cat README.md
The big one: the majority of my routine work now executes through AI agents (Claude Code, Codex, Ollama, Opencode) under a governance model I built – dedicated least-privilege identity, JIT elevation, read-only by default, human sign-off on anything privileged.
Four full rewrites of the same app: Pinecone that wouldn’t retrieve, Docker that wouldn’t run, and finally a CLI tool I actually use – full story on the blog.
A working Azure monitoring stack built almost entirely through conversations with Claude – including the hard-earned lessons about context loss and debugging loops. Write-up on the blog.
~$ cd personal/ && cat README.md
~$ ls courses/ | sort