Stanford CME295 Transformers & LLMs | Autumn 2025 | Lecture 7 - Agentic LLMs

For more information about Stanford’s graduate programs, visit: https://online.stanford.edu/graduate-education November 14, 2025 This lecture covers: • Retrieval-augmented generation • Advanced RAG techniques • Function calling • Agents • ReAct framework To follow along with the course schedule and syllabus, visit: https://cme295.stanford.edu/syllabus/ Chapters: 00:00:00 Introduction 00:06:38 RAG overview 00:27:39 Similarity search with SBERT and bi-encoders 00:34:25 Heuristic search with BM25 00:37:54 HyDE and contextual retrieval 00:41:00 Prompt caching 00:45:24 Re-ranking with cross-encoders 00:47:49 Retrieval evaluation with NDCG, MRR 00:59:28 Tool calling 01:26:22 Tool selection 01:29:17 Model Context Protocol (MCP) 01:31:56 Agents with ReAct 01:42:16 Safety and closing thoughts Afshine Amidi is an Adjunct Lecturer at Stanford University. Shervine Amidi is an Adjunct Lecturer at Stanford University. View the course playlist: https://www.youtube.com/playlist?list=PLoROMvodv4rOCXd21gf0CF4xr35yINeOy