About this course

This course introduces the foundations of AI-native data engineering for experienced data practitioners and explains how AI workloads reshape data architecture, pipeline design, platform responsibilities, and production expectations. Learners examine how traditional analytics pipelines evolve to support embeddings, vector retrieval, LLM integration, retrieval-augmented generation, semantic search, governance, observability, drift monitoring, and cross-team operational ownership. The course maps AI data lifecycle stages to practical data engineering work and explores design tradeoffs involving hosting, latency, freshness, access control, cost, and reliability. By the end of the course, learners will be able to redesign a legacy analytics-oriented pipeline into an AI-native architecture with documented tradeoffs, governance controls, observability considerations, and clear ownership boundaries.

Level: Intermediate  ·  Total duration: 15h 52m

What you'll learn

• Describe AI Native Data Engineering and explain how it differs from traditional data engineering.
• Explain core AI, ML, generative AI, LLM, embedding, vector database, and RAG concepts from a data engineering perspective.
• Map AI data lifecycle stages to practical data engineering responsibilities, including preparation, feature engineering, embedding, retrieval, evaluation, monitoring, governance, and drift management.
• Evaluate LLM integration choices, including prompting, hosting, cost, latency, scale, fine-tuning, and RAG.
• Design RAG and semantic search workflows that include source preparation, chunking, embeddings, vector retrieval, metadata, evaluation, governance, and access control.
• Redesign an analytics-oriented pipeline into an AI-native target architecture with documented tradeoffs, non-functional requirements, observability, governance, security, and ownership boundaries.

Skills you'll gain

• AI-native architecture (level 3)
• AI workload analysis (level 3)
• Architecture diagramming (level 3)
• AI stack component mapping (level 3)
• Source preparation (level 3)
• Embedding pipeline design (level 3)
• Retrieval integration (level 3)
• Prompting fundamentals (level 2)
• Hosting tradeoff analysis (level 3)
• Cost and latency evaluation (level 3)
• Chunking and metadata design (level 3)
• Vector retrieval (level 3)
• Governance and access control (level 3)
• Drift monitoring and observability (level 3)
• Architecture report writing (level 3)