Retrieval-Augmented Generation (RAG) is what lets an AI application answer questions using information it wasn't trained on, and this Mckay Wrigley course focuses on the retrieval half of that equation.
Core architecture
You'll learn how retrieval and generation fit together, then implement the retrieval strategies and generation techniques that connect the two into a system that actually performs well in practice.
Advanced search techniques
- Vector search for fast, relevant lookups
- Neural re-ranking to sharpen result quality
- Hybrid search combining multiple retrieval paradigms
Machine learning underpinnings
Across 22 lessons, the course also touches supervised and unsupervised learning, transformer models for language understanding, and reinforcement learning for systems that adapt over time — the ML tooling behind a RAG pipeline that's both accurate and efficient.
