Published Sep 1, 2023

710: LangChain: Create LLM Applications Easily in Python — with Kris Ograbek (@krisograbek)

Jon Krohn and Kris Ograbek delve into the creation of interactive LLM applications with the LangChain framework, exploring innovative data processing techniques like embeddings and data chunking, while also reflecting on the evolving art of podcast production and host transitions.
Episode Highlights
Super Data Science: ML & AI Podcast with Jon Krohn logo

Popular Clips

Episode Highlights

  • LangChain Intro

    introduces the LangChain framework, a powerful tool for developing large language model (LLM) applications. He shares his experience of creating a chatbot that interacts with podcast episodes, showcasing LangChain's capabilities in handling complex queries and providing insightful responses 1. Kris emphasizes the framework's potential by detailing a project where he used LangChain to enable users to chat with a podcast episode, demonstrating its practical applications 2.

    The idea came from two sources: deep learning AI courses and my desire to interact with content I consume.

    ---

    This innovative approach highlights LangChain's ability to transform how we engage with digital content 3.

       

    LLM Projects

    Building projects with LLMs requires a unique skill set, which Kris has developed through his work with Python and LangChain. He discusses the importance of specialization in LLMs, noting that combining prompt engineering with domain knowledge can create valuable applications 4. Kris believes that focusing on LLMs allows for the integration of diverse skills into something impactful.

    Large language models are a field where you can really combine prompt engineering and domain knowledge into something valuable.

    ---

    This approach not only enhances personal expertise but also contributes to the broader field of AI development 5.

       

    AI Tools

    Interactions with AI tools like GPT models have become integral to modern workflows. shares how these tools enhance productivity and creativity, describing them as "supercharged" assistants that simplify complex tasks 6. He highlights the transformative impact of AI on everyday activities, from drafting emails to developing machine learning models.

    It's magical. We now have, as data scientists, these state-of-the-art LLMs that make everything so easy.

    ---

    These interactions not only streamline processes but also foster a more positive and engaging work environment 7.

Related Episodes