# Building AI Agents with Databricks Day2 Lecture

> In this lecture, Eumar discusses the critical transition from demo to production for AI agents, emphasizing that 80% of GenAI projects fail to reach production due to issues like data quality and user trust. He explores essential technologies such as Databricks Model Serving, Guardrails, and MLflow…

- Web page: https://www.dataexpert.io/lesson/buildingaiagentswithdatabricksday2lecture-feb26-p736-l2035
- Program: [DataExpert](https://www.dataexpert.io/program/data-expert)
- Module: Week 6: Building AI Agents with Databricks
- Access: Requires enrollment in DataExpert
- Length: 1 h 3 min video
- Skills: MLOps, Data Modeling, Apache Spark, APIs, Databricks, Delta Lake
- Academy: DataExpert.io Academy

## About this lesson

In this lecture, Eumar discusses the critical transition from demo to production for AI agents, emphasizing that 80% of GenAI projects fail to reach production due to issues like data quality and user trust. He explores essential technologies such as Databricks Model Serving, Guardrails, and MLflow Evaluation, which can help ensure successful deployment. He highlights the importance of observability and monitoring in maintaining agent performance post-launch.

## Navigation

- Previous lesson: [Building AI Agents with Databricks Day1 Lab](https://www.dataexpert.io/lesson/buildingaiagentswithdatabricksday1lab-feb26-p736-l2034.md)
- Next lesson: [Building AI Agents with Databricks Day2 Lab](https://www.dataexpert.io/lesson/buildingaiagentswithdatabricksday2lab-feb26-p736-l2036.md)