Notes on AI agents, local models and making them work at work.https://bhavukmanocha.com/blog/https://bhavukmanocha.com/blog/build-your-own-instinct-with-hermes/https://bhavukmanocha.com/blog/build-your-own-instinct-with-hermes/Meta's Muse and the invite-only Instinct sell the same idea: an assistant that remembers you, lives in your chats and messages you first. Hermes Agent, an open-source agent from Nous Research, has every piece you need to build your own.Sun, 27 Sep 2026 00:00:00 GMThttps://bhavukmanocha.com/blog/plugging-claude-into-your-warehouse/https://bhavukmanocha.com/blog/plugging-claude-into-your-warehouse/Connect a model to the warehouse, ask a question, get a confident number. Anthropic's own data team says that setup creates "a false sense of precision". Their fix is mostly not about the model.Sun, 27 Sep 2026 00:00:00 GMThttps://bhavukmanocha.com/blog/qwen36-35b-on-an-8gb-gpu/https://bhavukmanocha.com/blog/qwen36-35b-on-an-8gb-gpu/Qwen3.6-35B-A3B has 36 billion parameters and still runs on a mid-range GPU at around 40 tokens a second. The trick is knowing which 7% of the model has to be on the card, and parking the rest in system RAM.Sun, 27 Sep 2026 00:00:00 GMT