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)
@hwchase17
-
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
-
Defining Memory in AI Applications: Beyond the Hype
By
–
The idea of memory is tantalizing, but also really vague What does it even mean for an application to have memory? Much like agents, there's a lot of hype and interest in this area, without a clear definition of what is actually means
-
Application-Specific Memory Implementation in AI Systems
By
–
We saw that applications that successfully implemented memory were implementing it in an application specific way A coding app may care about a users language, preferred libraries, and proficiency level A companion app may care about a users name, age, friends, etc
-
Vertical Memory Systems Rise Alongside Vertical AI Agents
By
–
In many ways… this is also similar to agents! Just as we are seeing the rise of **vertical** agents, we also think that we will see the rise of **vertical** memory
-
LangGraph Introduces Long-Term Memory Support for Agents
By
–
We're launching "long-term memory" support in LangGraph At its core, long-term memory is "just" a persistent document store that lets you *put*, *get*, and *search* for memories you've saved Why so simple?
-
LangGraph Launches Long-Term Memory Support for AI Agents
By
–
We put a lot of efforts into some great resources around this launch! Memory is a under explored and very open ended topic… more than anything, we wanted to share how we are thinking about it and get feedback Blog post: https://
blog.langchain.dev/launching-long
-term-memory-support-in-langgraph/
…
Conceptual docs: -
LLM Agency: How Agentic Applications Control Their Flow
By
–
The more agentic an application is, the more an LLM decides the control flow of the application
-
Building AI Agents with LangGraph and LangSmith
By
–
Interesting thread that mirrors how we see agents being built: 1. Break complicated task into a highly controllable domain specific workflow (LangGraph) 2. Test each component thoroughly (LangSmith)