By 2025, 80% of enterprises will integrate GenAI into production workflows, yet only 15% feel confident deploying reliable RAG systems. This Short Course was created to help Machine Learning and Artificial Intelligence professionals build, optimize, and evaluate production-grade GenAI applications on the Databricks platform. By completing this course, you'll be able to construct vector search pipelines from raw data, fine-tune models with MLflow tracking, and implement rigorous evaluation frameworks that ensure your GenAI systems meet real-world SLA requirements—skills you can apply immediately to customer-facing AI deployments.

By the end of this course, you will be able to: • Apply Databricks Lakehouse and vector search features to build a retrieval-augmented generation pipeline from raw data to queryable embeddings • Analyze fine-tuning experiment results in MLflow to select adapter parameters that balance output quality and latency constraints • Evaluate GenAI model responses for relevance, hallucination rate, cost, and latency, iterating prompt and context configurations to meet acceptance criteria This course is unique because it combines hands-on Databricks Lakehouse workflows with MLflow experiment tracking and production-grade evaluation metrics, bridging the gap between GenAI prototypes and enterprise deployments. To be successful in this course, you should have working knowledge of Python programming, basic machine learning concepts, and familiarity with cloud data platforms at the CB2 intermediate level.