Effective agent design patterns in production
summary
LlamaIndex is a Python/TypeScript framework with Llama Parse for cleaning PDFs/Word/PowerPoints, a paid Llama Cloud retrieval endpoint, a connector registry and prebuilt tools that integrate 400 models over 80 providers. Patterns proven in production: chaining, routing, parallelization (sectioning and voting), orchestrator-workers and evaluator-optimizers. Claim: layering agents on top of RAG raises accuracy and speed versus naive top-k RAG. Workflows support concurrency, visualizers, simple tool wrappers and one-line multi-agent composition that runs in real systems.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 1:08What is Llama Index
- 3:16What is an agent
- 5:45Why use RAG with agents
- 8:35Agent design patterns
- 15:21Q&A and resources
related talks
35mAdvanced RAG Techniques with LlamaIndex
32mBuilding Multi-Agent RAG Systems
01h 06mBuilding Smart Agents with LlamaIndex: A Comprehensive Overview
50mAI Agentic Workflows & Design Patterns
40mBuilding Effective AI Agents Simple, Composable Patterns
29mThe Future of AI Assistants
31mAgentic AI architecture patterns
56mHerding LLMs: Predictable Outcomes from Unpredictable Agents