Building Smart Agents with LlamaIndex: A Comprehensive Overview
summary
Shows a working Jupyter/React agent built with Llama 3.1 that actually calls tool plugins (calendar, weather API, finance, DuckDuckGo search) and falls back to search when an API is missing, with retries set to five and caching/rate‑limit recommendations for production. Demonstrations include a calculator call where multiply tool -> 20234567=9,239,041 and a live weather readout of Belgrade 7.2°C*, plus booking, flight search and stock lookup examples that complete end‑to‑end.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:24What is LlamaIndex
- 9:37Agents and tools
- 21:16How agents work
- 25:39Steps to build an agent
- 39:35Best practices for agents
- 43:07Demo overview
- 58:05Q&A session
related talks
16mEffective agent design patterns in production
41mBuilding LLM Agents – Challenges and Learnings
02h 08mIntroduction to AI Agents
42mUsing LlamaIndex for building Generative AI apps
39mTowards Useful, Reliable, and Scalable AI Agents
35mAdvanced RAG Techniques with LlamaIndex
09mBuilding LLM Agents with Tool Use
01h 28mBuilding LLM Agents Made Simple a