Introduction to the Databricks AI cookbook
The Databricks AI cookbook and its sample code take you from a proof-of-concept (POC) to a high-quality production-ready application using Mosaic AI Agent Evaluation and Mosaic AI Agent Framework on the Databricks platform.
The Databricks Generative AI Cookbook is a definitive how-to guide for building high-quality generative AI applications. High-quality applications are:
Accurate: They provide correct responses
Safe: They do not deliver harmful or insecure responses
Governed: They respect data permissions & access controls and track lineage
Developed in partnership with Mosaic AI’s research team, this cookbook lays out best-practice development workflow from Databricks for building high-quality RAG apps: evaluation-driven development. It outlines the most relevant ways to increase RAG application quality and provides a comprehensive repository of sample code implementing those techniques.
Tip
There are a few ways you can build a rag app using this cookbook:
You only have a few minutes and want to see a demo of Mosaic AI Agent Framework & Agent Evaluation.
You want to hop into code and deploy a RAG POC using your data.
You don’t have any data, but want to deploy a sample RAG application.
The Databricks approach to quality
Databricks takes the following approach to AI quality:
Fast, code-first developer loop to rapidly iterate on quality.
Make it easy to collect human feedback.
Provide a framework for rapid and reliable measurement of app quality.
This cookbook is intended for use with the Databricks platform. Specifically:
Mosaic AI Agent Framework that provides a fast developer workflow with enterprise-ready LLMops & governance.
Mosaic AI Agent Evaluation that provides reliable, quality measurement using proprietary AI-assisted LLM judges to measure quality metrics that are powered by human feedback collected through an intuitive web-based chat UI.
Code-based workflows
Choose the workflow below that most meets your needs:
Time required |
What you’ll build |
Link |
---|---|---|
10 minutes |
Sample RAG app deployed to web-based chat app that collects feedback |
|
2 hours |
POC RAG app with your data deployed to a chat UI that can collect feedback from your business stakeholders |
|
1 hour |
Comprehensive quality, cost, and latency evaluation of your POC app |