Single Prompt -> Multi Agent -> LangGraph This is a very common trend we see among agent developers Our most recent blog post with Rexera (real estate agent) highlights this The journey goes like this: Start with a single prompt, which: "struggled with the complexity of
@hwchase17
-
Five Sigma Deploys Claims Adjustment Agent in Production
By
–
Another day, another agent in production Excited to share this case study with Five Sigma on how they built and deployed an agent to adjust claims Some interesting nuggets:
How can a company launch an agent in a highly regulated space? Rigorous testing!
Good documentation -
Multi-agent workflows update: parallel tool use control
By
–
sneaky big update to support agent/multi-agent workflows disable_parallel_tool_use: i often want agents to make one (and only one) tool call consecutive user/assistant messages: often happens in multi-agent situations
-
Open Source PostgreSQL Implementation Available
By
–
i believe we have open source postgres implementtion
-
Building Account Qualification Agents: Cognitive Architectures Explained
By
–
Agents are here, you just have to know how to build them Excited to share this VERY detailed blogpost from @unifygtm on how they built their recent account qualification agent This contains all sorts of nuggets like:
The three different cognitive architectures they considered -
Custom AI Implementation Services for Enterprise Solutions
By
–
Because we think this is application specific, there's only so much we can do off the shelf… but we love working with companies in a more hands on manner to help them implement it! If that's interesting, please reach out!
-
LangGraph Memory Features: Cross-Thread Persistence Implementation
By
–
Conceptual video: https://
youtube.com/watch?v=JTL0yp
85FsE
… Conceptual docs: https://
langchain-ai.github.io/langgraph/conc
epts/memory/
… How-to docs: https://
langchain-ai.github.io/langgraph/how-
tos/cross-thread-persistence
… -
Memory Creation and Best Practices for AI Systems
By
–
Of course – there's still a lot of complexity in how you create memories to store here, and how you use them We believe this will be application specific, but we want share best practices and conventions, so we spent a ton of time writing conceptual docs on this (next tweet)
-
LangGraph Document Store for Memory Management
By
–
After a few iterations, we found the best way to do that was to build a document store into LangGraph This comes with:
– Flexible namespacing (to support multiple types of memory)
– JSON Document Storage
– Content-Based Filtering -
LangGraph Framework Enables Custom Cognitive Architectures Memory Management
By
–
With LangGraph, we created a low-level, highly flexible framework to give users the power to create custom cognitive architectures We wanted to give the same low-level control and flexibility to how users created and used memory