Agentic AI: Build a Multi-Agent Application with CrewAI
summary
Hands-on build using the open‑source Python CrewAI library: a two‑agent prompt optimizer that outputs a single enhanced prompt for a target LLM, a content generator wired to a flow that re‑runs until safety checks pass, and an anomaly detector using a conditional task that escalates when anomalies exceed a threshold (examples use >2 or >5). Tools shown: SerpDev web tool, multiple LLM providers, and Pydantic for deterministic outputs. Four plus one assignments provide ~80% starter code and the demos run end‑to‑end (API/Wi‑Fi hiccups noted).
AI-written summary · about
chapters (11)
- 0:00Introduction to workshop
- 10:48What is CrewAI
- 29:16Workshop setup and resources
- 42:09First assignment: Prompt optimization
- 1:00:28Solution to first assignment
- 1:11:58Second assignment: Content creation with guardrails
- 1:26:29Solution to second assignment
- 1:31:26Third assignment: Anomaly detection
- 1:41:17Solution to third assignment
- 1:48:42Fourth assignment: Onboarding system
- 1:52:14Q&A and wrap-up
related talks
01h 04mAgentic AI: Build a Multi-Agent Application with CrewAI
29mDemystifying AI Agents with Python Code
52mCut the Hype and Build Your Own AI Agents
41mGoing beyond copilot with AI Agents
01h 30mHow to build AI Agents: A step by step guide to creating AI agents
37m‘Nurturing’ AI: The Story of How an AI Copilot Was Built
49mAI Agents: Collaboration in Real-Life Applications
40mHow To Tame Your Agents?