Pycon Sweden 2024

Query your structured data with a LangChain AI agent
2024-11-15 , Auditorium

In this talk, we will explore the capabilities of LangChain, a versatile framework tailored for the development of applications powered by Large Language Models (LLMs).
We will see how to use LangChain to build advanced AI applications with autonomous AI agents.


In this talk, we will explore the capabilities of LangChain, a versatile framework tailored for the development of applications powered by Large Language Models (LLMs).
By bridging the gap between LLMs and sources of context, such as prompt instructions and few-shot examples, LangChain empowers applications to understand and reason about user queries effectively.
With over 90k stars on GitHub, LangChain stands as the leading Generative AI framework in terms of adoption.
During the presentation, attendees will witness a live demonstration showcasing LangChain's SQL agent in action. The agent will autonomously choose available tools to get information about the database schema, select relevant “golden sample” queries from a vector store that will reduce so-called hallucinations from the LLM, call the LLM to get a text-2-SQL answer, execute the query on the database, and build an answer from the query result to the initial question.

Christophe is a Senior Software Engineer at DataStax working on AI frameworks. Through his love for open source, he is also a LangChain Community Champion, an Apache Pulsar committer and a core team member of JHipster and OpenAPI-generator projects.