09-18, 09:00–10:30 (Europe/Amsterdam), Rokin Room - OBA Oosterdok
With this workshop you will familiarize yourself with the key concepts of an LLM agent using LangChain and the OpenAI chat completion endpoints.
LLM Agents are systems that use an LLM to reason through a problem and perform actions on the behalf of the user. Think about combining your LLM with an API that fetches the most recent weather information or real estate updates, and “teaching” it to decide when this API is relevant to use in a generated response. With this, Agents can perform various tasks, such as answering questions, generating text, or even engaging in conversation.
In this tutorial, we'll dive into the theory behind what agents are, how they function, and the scenarios in which they can be effectively used. With this workshop you will familiarize yourself with the key concepts of an LLM agent using LangChain and the chatgpt chat completion endpoints. At the end, you will have all the code you need for your very own agent and you will be able to build custom tools for your own use-case.
Repository: https://github.com/mkmbader/pydata_workshop_September2024
Requirements
- Python 3.8 or higher
- Jupyter notebook or jupyter-lab
Setting up your environment
The notebooks guide you how to set up the repo with Google Colab.
If you prefer to run it locally instead please follow the steps in the README.
Ana is a Data Scientist at the Amsterdam-based fintech Mollie. She works with ML and Gen AI to create solutions for automating and optimizing customer monitoring and payment processes. Ana enjoys exploring novel data approaches and practically implementing them to solve business challenges.
I am a Data Scientist, who is specialised in both, traditional machine learning and generative AI techniques. At Mollie, I am currently contributing to the development of MollieGPT, the company's chatbot. My background as a researcher in physics fuelled my love for unraveling hidden details in data and has equipped me with the unique ability to approach complex problems with a scientific mindset.