Evaluating agents One of the trickiest parts of building agents is evaluating agents We're tackling this with LangSmith. I'll sit down with Writer CTO Wassem Alshikh **tomorrow** in SF to talk about this See ya there! https://
lu.ma/x7ohaj2c
@hwchase17
-
Evaluating AI Agents with LangSmith and Writer CTO
By
–
-
Canvas UX Outperforms Chat for Document Work
By
–
Chat is great, but canvas ux is better for working with documents
-
Sick Canvas and Agent Experience Innovation
By
–
sick canvas + agent experience!!! https://t.co/HMmaj7amed
— Harrison Chase (@hwchase17) 19 mars 2025sick canvas + agent experience!!!
-
LangChain Customer Story: Rexera Implementation Success
By
–
heres a good customer story on trying out both https://
blog.langchain.dev/customers-rexe
ra/
… -
LangGraph Studio Enables Multi-Agent Distributed Systems Visualization
By
–
Multi agent systems with distributed subagents LangGraph Studio now supports visualizing and debugging these
-
Debugging Prompting Issues: Technical Solution Discussion
By
–
Probably a prompting issue. Ooc – can you say more about the use case? I can maybe hack something together
-
LangGraph Swarm: Multi-Agent Architecture with Handoff
By
–
I would look into langgraph swarm. Have one agent that is in change of getting the info, then hands off to another one that does the task:
-
LangGraph vs CrewAI: Low-Level Control and High-Level Frameworks
By
–
Langgraph is much lower level and more controllable. Crewai is higher level and off the shelf. We have some prebuilt packages that are built on top of langgraph that are similar to crew, like langgraph supervisor:
-
Canvas UX Becomes Standard for LLM Document Interaction
By
–
Canvas ux is becoming the standard for interacting with LLMs on documents Want to implement yourself? Check out open canvas! https://
github.com/langchain-ai/o
pen-canvas
… -
Low Code Agent Builder Debate and Criticism
By
–
And you were trying to argue for a low code agent builder??