Debugging agents has never been more fun
@hwchase17
-
Human Annotations Critical for LLM Training Data Flywheel
By
–
The importance of human annotations Human annotations on LLM outputs/trajectories are crucial for getting the data flywheel turning Decagon called this out as a key component of their AI agent platform. Learn more about it NEXT TUESDAY https://
lu.ma/w7y0bqwr -
LangGraph State Management Primitives Now Accessible Beyond Framework
By
–
excited to make low level primitives for state management accessible outside of LangGraph next up: higher level, off-the-shelf implementations for short and long term memory
-
AI Agent Flywheel Architecture: Core Components Explained
By
–
What does a flywheel for AI agents look like? According to Decagon (leading customer support agent) it involves:
Core AI Agent
Routing
Agent Assist
Admin Dashboard
QA Interface Want to learn more? We're doing a fireside chat next TUESDAY: https://
lu.ma/w7y0bqwr -
COT Benchmarking Training in Modern AI Models
By
–
these were easy to benchmark. COT would be easy as well but is trained into most models these days so i dont think it would change too much
-
Framework-Agnostic AI Optimizers vs DSPy Dependency
By
–
dspy more powerful, but requires using the dspy framework (in addition to their optimizers) i think framework-agnostic optimizers are more usable
-
Production-ready prompt optimization methods rollout begins
By
–
We're going to start rolling out production ready versions of some of these prompt optimization methods – if interested in early access, fill out form here: https://
docs.google.com/forms/d/e/1FAI
pQLSdK1pZihqohabtGRQ99LZ2Rdwo5dnpnzJrBFRdojVSFq7k8eg/viewform?usp=dialog
… -
Claude-Sonnet Outperforms o1 for Meta-Prompting and Optimization
By
–
I think a lot of people are sleeping on using claude-sonnet to do meta-prompting/prompt optimization we found its better than o1 (and cheaper/faster) it still struggles for complex tasks (curious to see how o3 would do) but it works quite well for simpler one
-
LangChain Community Building Through In-Person Events
By
–
Great to see the LangChain community organizing irl events LangChain was started by going to a bunch of meetups and hearing what people were building – lots of value in these types of events