@max_grev azure foundry for OA models?
those have always had a higher latency than any other inference provider https://t.co/92hFfwigUH (for some models is 2x what OA gives u)
@chronark@NotionHQ i got some repo mapping diagrams in an onboarding guide and has worked pretty well.
kind of like treating the agent as a new eng getting their hands on the codebase.
tbf i do mostly use astra for these kind of workflows. grok doesn't perform as well for these kind of activities
JavaScript was never designed for the server. If that doesn't change soon, JavaScript on the server will be replaced by Rust.
The fix needs to happen in the engine. Sound types. Ahead-of-time compilation. Threads with shared objects.