Keep pulling the thread on Fully Connected Tokyo.
The LangChain framework is used for building AI applications, including those involving AI agents and processing chat messages.
AI applications built with LangChain can incorporate capabilities such as function calling, semantic search, and handling FAQs.
AI agents can be integrated with Robotic Process Automation (RPA) to automate business workflows.
LangGraph is a component used with the LangChain framework for building stateful, multi-actor AI applications.
A multi-provider strategy for large language models is recommended, utilizing models from OpenAI, Google (Gemini), and Anthropic (Claude).
Feedback for AI systems can be managed through a process of human annotation using a tool called Weave.