Follow the blog
Paste this address into any feed reader — Feedly, Inoreader, NetNewsWire, Readwise Reader — and new posts arrive automatically.
https://joessattes.github.io/Sattaya-AI-Website/rss.xmlOpen in Feedly Open in Inoreader
Latest
- Reading a model's architecture from its safetensors header, without downloading the weights
- Why emotion models fall apart outside the lab (and what actually helps)
- FinOps-Estimator: sizing GPUs for a voice AI pipeline
- Fitting an LLM in your own server room: the memory math behind on-prem AI
- TurboQuant, reviewed: when a theory paper walks straight into production