Writing about RAG performance, Unity ↔ React bridges, paywall design, and the weirder corners of full-stack work. One entry per week when I'm being honest, one per month when I'm not. Personal log, Stardate 78xxx.x.
A covering index sat right there, and Postgres chose a seq scan anyway. My friend called it a hallucination. I rebuilt the experiment with 5M rows — the planner wasn't hallucinating, it was doing arithmetic.
The advice says NoSQL for low latency. But why? Following the questions that actually answer it — write paths, sharding, joins, and why MongoDB's $lookup doesn't save you.