joined @LangChain almost three months ago and have spent a ton of time thinking about evals! i think that harbor is the future of eval frameworks so really enjoyed being able to bring it into tighter integration with our stack!!
🎉 So excited about dynamic subagents!
Dynamic subagents give your agent the ability to dynamically write code and programmatically spawn subagents. This dramatically helps your agent tackle large tasks at scale!
Go check out our blog or watch the demo video to learn more about this!
this reply just got me thinking. does the future look like frequently switching models per-task?
- ModelA for coding
- ModelB for writing
- ModelC for posting memes
Not serious about the last one ofc but..... I think to think more about this