// blogs · index

Things I'm writing.

Notes from production — engineering tradeoffs, AI systems, and the boring craft behind shipping software that actually works. New posts when I have something worth saying.

1 published~monthly

2026 · 04
Cutting LLM inference cost by 90%
Prompt caching, async orchestration, and the boring engineering that makes GenAI affordable.
8 min

Want a heads-up on new posts?

No newsletter yet — just follow me on socials.

twitter linkedin email me

Things I'm writing.

Cutting LLM inference cost by 90%

Want a heads-up on new posts?