Skip to main content

Generative AI Engineering with Databricks

This course is aimed at data scientists, machine learning engineers, and other data practitioners who want to build generative AI applications using the latest and most popular frameworks and Databricks capabilities. 


Note: Databricks Academy is transitioning to a notebook-based format for classroom sessions within the Databricks environment, discontinuing the use of slide decks for lectures. You can access the lecture notebooks in the Vocareum lab environment.


Below, we describe each of the four, four-hour modules included in this course.

Building RAG Agents with Agent BricksThis course provides hands-on training for building retrieval agents using Databricks Agent Bricks. Participants will learn to explore and query Knowledge Assistants, parse unstructured documents into structured data using AI Functions, chunk text for semantic retrieval, build Vector Search indexes, and create production-ready Knowledge Assistants backed by multiple knowledge sources.


Building Agentic Applications on Databricks: This course teaches students how to build production-grade agentic applications on Databricks. Students learn to create governed agent tools using Unity Catalog and MCP, build single and multi-agent systems with the OpenAI Agents SDK, and leverage Agent Bricks and Genie for knowledge-assistant use cases orchestrated with a supervisor agent. The course covers the full progression from tool prototyping to production deployment, with hands-on experience using MLflow tracing to observe agent execution.


Agent Evaluation on DatabricksThis course teaches students how to systematically evaluate AI agents using MLflow's evaluation framework, addressing the unique challenges of non-deterministic AI systems that traditional software testing cannot handle. Students learn to implement various evaluation approaches including built-in judges for common criteria like correctness and safety, guideline judges for business-specific requirements, and custom judges for specialized needs. The course covers both offline evaluation using curated datasets and online production monitoring, with hands-on experience using MLflow's tracing capabilities to understand agent execution patterns and collect human feedback from different stakeholder types. Through practical demonstrations and labs, students develop skills in creating evaluation workflows that drive continuous quality improvements throughout the AI agent development lifecycle.


Deploying and Monitoring Agent Applications on DatabricksThis course covers the end-to-end lifecycle for deploying and monitoring generative AI agents on Databricks. Participants will learn how to deploy agents as Databricks Apps using Declarative Automation Bundles (DABs), integrate tools via the Model Context Protocol (MCP), instrument agents with MLflow Tracing, and evaluate production quality using scorers, multi-turn judges, and online evaluation. Through hands-on demos and labs, participants will gain practical experience building, observing, and monitoring production-grade AI agents on the Databricks platform.

Skill Level
Associate
Duration
16h
Prerequisites

The content was developed for participants with these skills/knowledge/abilities:  


1. Building RAG Agents with Agent Bricks

• Basic SQL knowledge for querying tables and using built-in functions

• Basic Python programming experience

• Familiarity with the Databricks workspace UI (e.g., navigating Unity Catalog)

• Understanding of Unity Catalog concepts including catalogs, schemas, and volumes

• Familiarity with fundamental LLM concepts

• Understanding of structured vs unstructured data


2. Building Agentic Applications on Databricks

Python-Specific Skills:

• Basic Python syntax and data structures

• Understanding of functions, classes, and decorators

• Experience with Python package management and imports

• Familiarity with JSON data handling

• Basic understanding of async/await patterns


SQL-Specific Skills:

• Basic SQL query syntax (SELECT, FROM, WHERE)

• Understanding of SQL functions and user-defined functions

• Experience with Unity Catalog SQL functions


Databricks-Specific Skills:

• Understanding of Databricks workspace navigation and notebook interface

• Knowledge of Unity Catalog structure (catalogs, schemas, tables, volumes, functions)

• Experience with Databricks compute resources and serverless computing

• Familiarity with MLflow experiment tracking

• Understanding of Databricks model serving endpoints


GenAI/Agent-Specific Skills:

• Basic understanding of LLMs and their capabilities

• Knowledge of prompt engineering and system prompts

• Familiarity with tool-calling agents and function calling concepts

• Basic awareness of the Model Context Protocol (MCP)


3. Agent Evaluation on Databricks

• Intermediate Python programming experience

• Basic SQL knowledge for querying and creating functions

• Familiarity with Databricks Data Intelligence Platform

• Understanding of Unity Catalog concepts including catalogs and schemas

• Basic understanding of large language models (LLMs) and prompt engineering

• Basic knowledge of MLflow


4. Deploying and Monitoring Agent Applications on Databricks

• Familiarity with Databricks workspace and notebooks

• Familiarity with Unity Catalog

• Experience building agents using the OpenAI Agents SDK

• Basic knowledge of MLflow and Python

• Familiarity with GenAI agent concepts (LLM calls, tool invocation, retrieval)

Outline

1. Building RAG Agents with Agent Bricks

Introduction to RAG and Agent Bricks Knowledge Assistant

• Agent Bricks for Retrieval and Context Engineering

• Demo: Exploring the Knowledge Assistant

Building Document Parsing Pipelines on Databricks

• Document Parsing and Chunking Strategies

• Demo: Transforming PDFs to Structured Data

Vector Search and Knowledge Assistants

• AI Search on Databricks

• Demo: Chunking PDFs and Vector Search

• Lab: Creating and Curating A Knowledge Assistant


2. Building Agentic Applications on Databricks

Foundations

• Agents, MCP, and AI Governance on Databricks

• Demo: Building Agent Tools on Databricks

• Lab: Assessing and Fixing Agent Tools

Agent Development

• Building Agents with the OpenAI Agents SDK and MLflow

• Demo: Building Single Agents with the OpenAI Agents SDK

• Demo: Multi-Agent Orchestration with the OpenAI Agents SDK

Agent Bricks and Beyond

• Agent Bricks and Genie

• Lab: Building a Supervisor Agent with Agent Bricks

• BONUS: Building Single Agents with LangChain


3. Agent Evaluation on Databricks

AI Agent Evaluation Fundamentals

• The Challenge of Evaluating AI Agents

• Demo: Agent Setup

• MLflow's Evaluation Framework

AI Agent Evaluation Fundamentals

• Built-In Judges

• Demo: Using MLflow Built-In Judges

• Guideline Judges

• Demo: Guideline Judges with MLflow

• Custom Judges and Feedback

• Demo: Custom Judges with MLflow

• Lab: Applying Agent Evaluation

Custom Judges and Human Feedback

• Offline vs. Online Evaluation Strategies

• Lab: Developer and SME Feedback with MLflow


4. Deploying and Monitoring Agent Applications on Databricks

Agent Deployment

• Agent Deployment on Databricks

• Demo: Deploying an Observable Agent

Tool Integration and Observability

• Tool Integration and Observability

• Demo: Tracing for Production Agents

Production Evaluation and Monitoring

• Production Evaluation and Monitoring

• Lab: Online Evaluation with Multi-Turn Conversations

• Lab: Backfilling and Archiving Multi-Turn Conversations

Upcoming Public Classes

Date
Time
Your Local Time
Language
Price
Jul 28 - 31
11 AM - 03 PM (Asia/Singapore)
-
English
$1500.00
Jul 28 - 31
01 PM - 05 PM (Europe/London)
-
English
$1500.00
Jul 30 - 31
09 AM - 05 PM (Europe/London)
-
English
$1500.00
Aug 19 - 20
09 AM - 05 PM (America/New_York)
-
English
$1500.00
Aug 19 - 20
09 AM - 05 PM (America/Los_Angeles)
-
English
$1500.00
Aug 25 - 26
09 AM - 05 PM (Europe/Paris)
-
English
$1500.00
Sep 08 - 09
09 AM - 05 PM (Asia/Singapore)
-
English
$1500.00
Sep 09 - 10
09 AM - 05 PM (America/New_York)
-
English
$1500.00
Sep 29 - 30
09 AM - 05 PM (Europe/Paris)
-
English
$1500.00
Sep 29 - 30
09 AM - 05 PM (America/Chicago)
-
English
$1500.00
Oct 13 - 14
09 AM - 05 PM (Asia/Singapore)
-
English
$1500.00
Oct 14 - 15
09 AM - 05 PM (America/Los_Angeles)
-
English
$1500.00
Oct 20 - 21
09 AM - 05 PM (Europe/London)
-
English
$1500.00
Oct 27 - 30
10 AM - 02 PM (Asia/Kolkata)
-
English
$1500.00

Public Class Registration

If your company has purchased success credits or has a learning subscription, please fill out the Training Request form. Otherwise, you can register below.

Private Class Request

If your company is interested in private training, please submit a request.

See all our registration options

Registration options

Databricks has a delivery method for wherever you are on your learning journey

Runtime

Self-Paced

Custom-fit learning paths for data, analytics, and AI roles and career paths through on-demand videos

Register now

Instructors

Instructor-Led

Public and private courses taught by expert instructors across half-day to two-day courses

Register now

Learning

Blended Learning

Self-paced and weekly instructor-led sessions for every style of learner to optimize course completion and knowledge retention. Go to Subscriptions Catalog tab to purchase

Purchase now

Scale

Skills@Scale

Comprehensive training offering for large scale customers that includes learning elements for every style of learning. Inquire with your account executive for details

Upcoming Public Classes

Generative AI Engineer

Generative AI Engineering with Databricks - Spanish

Este curso está dirigido a científicos de datos, ingenieros de machine learning y otros profesionales de datos que desean crear aplicaciones de IA generativa utilizando los frameworks más recientes y populares y las funcionalidades de Databricks.

Nota: Databricks Academy está migrando a un formato basado en notebooks para las sesiones en el aula dentro del entorno de Databricks, y deja de utilizar diapositivas para las clases. Puede acceder a los notebooks de las clases en el entorno de laboratorio de Vocareum.

A continuación, describimos cada uno de los cuatro módulos de cuatro horas incluidos en este curso.

Building RAG Agents with Agent Bricks: este curso ofrece capacitación práctica para crear agentes de recuperación utilizando Databricks Agent Bricks. Los participantes aprenderán a explorar y consultar Knowledge Assistants, analizar documentos no estructurados para convertirlos en datos estructurados mediante AI Functions, dividir texto en fragmentos (chunks) para la recuperación semántica, crear índices de Vector Search y crear Knowledge Assistants listos para producción respaldados por múltiples fuentes de conocimiento.

Building Agentic Applications on Databricks: este curso enseña a los estudiantes a crear aplicaciones agénticas de nivel de producción en Databricks. Los estudiantes aprenden a crear herramientas de agente gobernadas mediante Unity Catalog y MCP, crear sistemas de agente único y multiagente con el OpenAI Agents SDK, y aprovechar Agent Bricks y Genie para casos de uso de knowledge assistant orquestados con un agente supervisor. El curso abarca toda la progresión, desde la creación de prototipos de herramientas hasta la implementación en producción, con experiencia práctica utilizando el rastreo (tracing) de MLflow para observar la ejecución del agente.

Agent Evaluation on Databricks: este curso enseña a los estudiantes a evaluar sistemáticamente agentes de IA utilizando el framework de evaluación de MLflow, abordando los desafíos únicos de los sistemas de IA no deterministas que las pruebas de software tradicionales no pueden manejar. Los estudiantes aprenden a implementar diversos enfoques de evaluación, incluidos los jueces integrados (built-in judges) para criterios comunes como la corrección y la seguridad, los jueces basados en directrices para requisitos específicos del negocio y los jueces personalizados para necesidades especializadas. El curso abarca tanto la evaluación sin conexión utilizando conjuntos de datos (datasets) seleccionados como el monitoreo de producción en línea, con experiencia práctica utilizando las capacidades de rastreo (tracing) de MLflow para comprender los patrones de ejecución del agente y recopilar retroalimentación humana de diferentes tipos de partes interesadas. A través de demostraciones y laboratorios prácticos, los estudiantes desarrollan habilidades para crear flujos de trabajo de evaluación que impulsan mejoras continuas de calidad a lo largo del ciclo de vida de desarrollo de los agentes de IA.

Deploying and Monitoring Agent Applications on Databricks: este curso abarca el ciclo de vida completo para implementar y monitorear agentes de IA generativa en Databricks. Los participantes aprenderán a implementar agentes como Databricks Apps utilizando Declarative Automation Bundles (DABs), integrar herramientas mediante el Model Context Protocol (MCP), instrumentar agentes con MLflow Tracing y evaluar la calidad en producción utilizando scorers, jueces de múltiples turnos y evaluación en línea. A través de demos y laboratorios prácticos, los participantes adquirirán experiencia práctica en la creación, observación y monitoreo de agentes de IA de nivel de producción en la plataforma Databricks.

Paid
16h
Lab
instructor-led
Associate
Generative AI Engineer

Generative AI Engineering with Databricks - French

Ce cours s’adresse aux data scientists, aux ingénieurs en machine learning et aux autres professionnels des données qui souhaitent créer des applications d’IA générative à l’aide des frameworks les plus récents et les plus populaires et des fonctionnalités de Databricks.

Remarque : Databricks Academy passe à un format basé sur des notebooks pour les sessions en classe dans l’environnement Databricks, et n’utilise plus de diaporamas pour les cours. Vous pouvez accéder aux notebooks de cours dans l’environnement de laboratoire Vocareum.

Ci-dessous, nous décrivons chacun des quatre modules de quatre heures inclus dans ce cours.

Building RAG Agents with Agent Bricks : ce cours propose une formation pratique à la création d’agents de récupération à l’aide de Databricks Agent Bricks. Les participants apprendront à explorer et à interroger les Knowledge Assistants, à analyser des documents non structurés pour les convertir en données structurées à l’aide des AI Functions, à découper le texte en fragments (chunks) pour la récupération sémantique, à créer des index Vector Search et à créer des Knowledge Assistants prêts pour la production s’appuyant sur plusieurs sources de connaissances.

Building Agentic Applications on Databricks : ce cours enseigne aux étudiants comment créer des applications agentiques de niveau production sur Databricks. Les étudiants apprennent à créer des outils d’agent gouvernés à l’aide d’Unity Catalog et de MCP, à créer des systèmes à agent unique et multi-agents avec l’OpenAI Agents SDK, et à exploiter Agent Bricks et Genie pour des cas d’usage de knowledge assistant orchestrés avec un agent superviseur. Le cours couvre l’ensemble de la progression, du prototypage des outils au déploiement en production, avec une expérience pratique de l’utilisation du traçage (tracing) MLflow pour observer l’exécution des agents.

Agent Evaluation on Databricks : ce cours enseigne aux étudiants comment évaluer systématiquement les agents IA à l’aide du cadre d’évaluation de MLflow, en relevant les défis propres aux systèmes d’IA non déterministes que les tests logiciels traditionnels ne peuvent pas gérer. Les étudiants apprennent à mettre en œuvre diverses approches d’évaluation, notamment les juges intégrés (built-in judges) pour des critères courants tels que l’exactitude et la sécurité, les juges basés sur des directives pour des exigences propres à l’entreprise, et les juges personnalisés pour des besoins spécialisés. Le cours couvre à la fois l’évaluation hors ligne à l’aide d’ensembles de données (datasets) sélectionnés et la surveillance de la production en ligne, avec une expérience pratique de l’utilisation des capacités de traçage (tracing) de MLflow pour comprendre les schémas d’exécution des agents et recueillir les retours humains de différents types de parties prenantes. Grâce à des démonstrations et des laboratoires pratiques, les étudiants développent des compétences pour créer des flux de travail d’évaluation qui favorisent l’amélioration continue de la qualité tout au long du cycle de vie de développement des agents IA.

Deploying and Monitoring Agent Applications on Databricks : ce cours couvre le cycle de vie de bout en bout pour le déploiement et la surveillance des agents d’IA générative sur Databricks. Les participants apprendront à déployer des agents en tant que Databricks Apps à l’aide des Declarative Automation Bundles (DABs), à intégrer des outils via le Model Context Protocol (MCP), à instrumenter les agents avec MLflow Tracing, et à évaluer la qualité en production à l’aide de scorers, de juges multi-tours et de l’évaluation en ligne. Grâce à des démos et des laboratoires pratiques, les participants acquerront une expérience pratique de la création, de l’observation et de la surveillance d’agents IA de niveau production sur la plateforme Databricks.

Paid
16h
Lab
instructor-led
Associate

Questions?

If you have any questions, please refer to our Frequently Asked Questions page.