fun fact, langchain has a super high ratio as well, despite a very high star count
@hwchase17
-
LangGraph: Production-Ready Agent Framework Low Level Control
By
–
Because langgraph is a production ready agent framework… it’s very low level, highly controllable, no hidden prompts Helps you get to production without getting in the way, so high usage But a bit of a steeper learning curve (we’re working on that!) so less GitHub stars
-
LangGraph Feature Discussion: Built-in Capabilities Analysis
By
–
What exactly do you think it would improve about langgraph? We’ve already had this built in for a while
-
LangGraph Persistence Layer: Universal Agent Framework Integration
By
–
we've spent a lot of time on the langgraph persistence/state layer enables production ready short term & long term memory, checkpointing, human-in-the-loop, time travel, etc if we made this work with ANY agent framework (including DIY) would that be interesting to folks?
-
Agents in Production: 2024 Breakthrough Year for LangGraph
By
–
2024 is the year agents started to work It's still not easy, and they're vertical, specific, controllable agents… but Replit, Uber, LinkedIn, Elastic, Appfolio (and many more) have already shared their stories of building agents with LangGraph https://
blog.langchain.dev/top-5-langgrap
h-agents-in-production-2024/
… -
TypedDict and Data Classes for LangGraph State Management
By
–
We mainly try to use TypedDict to structure LangGraph state/LLM responses, data classes are nice for post init work
-
Human-in-the-loop and multi-agent architectures key to agent development
By
–
We've chatted with Replit a few times about their agent Both times they've mentioned human-in-the-loop and multi-agent architectures as being key pieces to building their agent I expect both of these will be key in building agents in the future!
-
Elastic AI Assistant Evolution: From RAG to Agentic Capabilities
By
–
Has been cool to see Elastic AI Assistant evolve from RAG to be more agentic over the past year I particularly like their work on "ES|QL query generation" – using LLMs to generate a particular DSL is a pretty good use case imo
