Mastering LLMs: From Prompt Engineering to Agentic AI
summary
Covers concrete workflows: prompt tricks like few‑shot, chain‑of‑thought and ReAct for grounding; RAG implemented as tool calls with both vector versus lexical retrieval and practical chunking; demos include a Python chatbot and a DSPY‑driven Shopify emailer that run (back ends faked but pattern works); surprising context shifts from 248‑token window to 100k+ / million‑token models and a caution that fine‑tuning is costly and brittle so prompt‑engineering, tool design, and automated LLM‑as‑judge evals often decide success.
AI-written summary · about
chapters (8)
- 0:00Large language models intro
- 2:13Prompt engineering basics
- 13:29Retrieval augmented generation
- 24:35Agentic AI and conversational agents
- 43:09Workflow agency and multi-step tasks
- 1:06:48Evaluating LLM apps
- 1:23:03Fine-tuning strategies
- 1:28:59Q&A
related talks
26mMastering Large Language Models: Prompt Engineering for Precision & Performance
19mUnlocking the Full Potential of LLMs: Advanced Prompt Engineering Techniques
26mHow To Win Friends & Influence LLMs (with Prompt Engineering)
59mTutorial: How to Win Friends & Influence LLMs (with Prompt Engineering)
36mLLMs and AI agents demystified
33mOpening Keynote: LLMs, between fantasy and reality - the poin...
48mBeyond the Prompt: Evaluating, Testing, and Securing LLM Applications
35mBuilding LLM-based Agents