Carryover: Portable state handoff for LLMs
I built a a portable work state CLI tool to easily switch between LLMs
Read01
Essays
When Small Models Are Good Enough: A Peek into Geometry of Intelligence
Let's take a local manifold walk to understand why small models are sometimes better, more resource responsible and won't burn a hole in your pocket.
ReadClaude Code Rate Limits: Diagnosing my token usage
I'm running into rate limits every hour or so into using claude code. I analyzed one session: < 3% of my AI bill went to intelligence. The rest? Regurgitating stale and bloated context
ReadStrange Habits of LLMs: Oops I Did It Again
I was fixing a bug for my newborn's app and ChatGPT had a complete brain-body disconnect: It kept reasoning 'don't do the thing' → did it → 'oops I did this' → redo → 'oops I did it again'
ReadBuild Your Own Glean: A Production RAG System
Demos are easy. Production is hard. If you want internal knowledge search that actually works, you're not 'adding RAG'—you're building a pipeline where the LLM is the easiest part.
ReadAI Evals: What I learned shipping production grade billing systems
From Evals 101 to Production Grade systems.
ReadRebuilding (resuscitating) this site
Product lessons: atomic desires, two way doors, and future complexity backseat
Read