2026
Snake
A browser-based Snake game published with GitHub Pages.
Posts
View all →
OpenAI Just Admitted the Boring Part Is the Product
OpenAI's new Deployment Company is not another model launch. It is a bet that enterprise AI will be won by the teams who can wire models into messy real workflows.
Benchmarks Are Thermometers, Not Report Cards
LLM benchmarks are useful when you treat them like instruments, not trophies. Here is how to read MMLU, Arena, SWE-bench, HELM, and your own evals without turning the leaderboard into a religion.
Google Finance Put AI on the Ticker Tape
Google's AI-powered Finance experience is expanding to 100+ countries. The useful part is faster research; the trap is mistaking a clean interface for a clean answer.
The Agent Didn't Delete Your File. It Sanded It Down.
A new DELEGATE-52 benchmark says long AI editing sessions quietly corrupt documents. The useful lesson is not 'never delegate' — it is 'make every edit inspectable.'