kadhirvelm

Wonder how open source models have been affecting their growth. As a consumer, we’ve started cracking down on reducing token expenditure lately. Not terribly hard to get similar value for less right now

We’ve been working on exactly this, deriving documentation from external systems (like GitHub, etc) into a giant set of docs that agents can reference and edit. Works way better than I would’ve originally expected. Suddenly these things are able to reference granola notes, a slack discussion, and an RFC when making coding decisions. Super helpful in a lot of unexpected ways!

What do people think of apps switching to a lower tier of model if your $$ threshold was exceeded?

Every company now has thousands of agents trying to get work done, and almost none of them know what the organization actually knows. To tackle this problem, we built a context engine.