Building Retrieval Augmented Generation (RAG)
Building Retrieval Augmented Generation (RAG)
COM SCI 910.2
This course teaches students to design and deploy Retrieval-Augmented Generation systems, combining LLMs with external data for accurate, scalable AI applications using modern tools, evaluation frameworks, and cloud platforms.
Get More Info
What you can learn.
What you can learn.
Design and implement end-to-end RAG architectures.
Select and optimize embedding models for different use cases.
Build scalable vector search systems.
Evaluate RAG system performance across multiple dimensions.
Deploy RAG applications in production environments.
Troubleshoot and optimize RAG system performance.
About This Course
This hands-on course introduces students to the design and development of Retrieval-Augmented Generation (RAG) systems — powerful architectures that combine large language models (LLMs) with external knowledge sources to improve accuracy, reduce hallucinations, and enhance domain-specific reasoning. Students progress from foundational understanding to real-world implementation, gaining experience with the latest tools and frameworks in the RAG ecosystem. Through guided labs and projects, learners will build complete, production-ready RAG pipelines — from data ingestion and embedding optimization to retrieval tuning, evaluation, and deployment on cloud platforms. The course emphasizes both engineering depth and practical evaluation, ensuring students understand the trade-offs between model quality, latency, and cost. By the end of the course, students will have developed and deployed a portfolio-ready RAG application with end-to-end documentation, demonstrating their ability to integrate vector databases, optimize retrieval strategies, implement automated evaluation using frameworks like RAGAS and TruLens, and containerize applications for scalable production environments.