Deploy LLM Chatbots with RAG and Databricks Vector Search

What you’ll learn

In this product tour, you'll see how an end-to-end RAG (Retrieval Augmented Generation) system works on Databricks and enhances your AI application's accuracy and relevance, by dynamically pulling in the most current and pertinent information for each query.  See how you can seamlessly integrate RAG into your workflows, offering an intuitive user experience that simplifies complex data interactions, ensuring your team can focus on strategic tasks rather than data management. In this tour you'll learn about:

  • Setting up a workflow to ingest unstructured data (PDFs) and save them into Delta tables
  • Using an embedding model to transform text data into vectors and store them into a vector database
  • Serving Embedding Models, Foundational Language Models, and even langchain chains!
  • Chaining your LLM together with your data to augment the model's responses  

If you want to try this in your own workspace, check out the product tutorial.

Launch product tour

Recommended

datascience icon

Tutorial

Build High-Quality RAG Apps with Mosaic AI Agent Framework and Agent Evaluation, Model Serving, and Vector Search

data science

On-Demand Video

Lakehouse Monitoring and Vector Search

demothumb-small-tour-devexperience.png

Product Tour

Discover LakehouseIQ: The AI-Powered Engine That Uniquely Understands Your Business