benmann

I welcome this release. Lots of good reasons to have these models and practices in place, even outside regulated industries. Even the EU AI act make some of this encessary (in theory). I've built redaction and rehydration through specilized NER models into https://grepture.com, so definitely adding this to the pipeline. Optionally sitting in the hot path, allowing actual tinkering with requests before/after they hit the LLM can be really handy for compliance or dircet user input scenarios.

You're not wrong, this is an issue! It's far from ideal, but it's also not unusal. Probably depends on team size and seniority as well.

Yes, exactly! We're planning to add more databases and other data sources (APIs) in the future, but with a strong focus on Supabase.

Hey, I'm the creator of Ummm. Not really on the roadmap anytime soon, but it's also still early for the product! Right now it's more about getting a grasp of what people want out of it, squashing some bugs and improving the UX further — before moving development further into a specific direction.