Agentic AI & Multi-Agent Systems
Agentic AI & Multi-Agent Systems on AI-ML Companion: Build autonomous AI systems with tool use and agentic workflows. 34 interactive modules with live visualizations, quizzes, and hands-on Python coding.
Start free: Introduction to AI Agents, Autonomous AI Research, Project: Smart Claims Processor are fully open to everyone, no account required. The other 31 modules are part of AI-ML Companion Premium; every title and summary is listed below so you can see exactly what the track covers before deciding.
Modules in this track
- Introduction to AI Agents (free) - What are agents and why they matter
- Agent Lifecycle: The Complete Architecture (premium) - The 6-phase lifecycle powering every AI agent
- Function Calling (premium) - OpenAI and Anthropic tool use APIs
- Structured Outputs (premium) - JSON mode, schemas, and reliable parsing
- Designing Tools (premium) - Creating effective tools for agents
- Agent Tool Ecosystem (premium) - Browser, code, API, and database tools for agents
- Computer-Use Agents (premium) - Driving a real screen: pixels, grounding, verification, and containment
- ReAct Pattern (premium) - Reasoning and acting in loops
- Agent Planning (premium) - Task decomposition and planning strategies
- Reasoning Patterns (premium) - Chain-of-Thought, Self-Consistency, and structured reasoning
- Tree of Thought (premium) - Branching exploration, parallel reasoning, and backtracking
- LangChain Fundamentals (premium) - Building agents with LangChain
- Agent Frameworks (premium) - CrewAI, AutoGen, and framework comparison
- Vector Databases (premium) - Pinecone, Weaviate, ChromaDB for RAG
- Advanced RAG (premium) - Chunking strategies, hybrid search, reranking
- Agent Evaluation (premium) - Testing and evaluating agent performance
- Model Context Protocol (premium) - MCP for standardized tool interfaces
- Agent Memory Systems (premium) - Short-term, long-term, and episodic memory for agents
- LangGraph Agents (premium) - Stateful multi-step agents with LangGraph
- Multi-Agent Patterns (premium) - Orchestration, delegation, and collaboration patterns
- CrewAI Multi-Agent Systems (premium) - Role-based crews and collaborative AI teams
- OpenAI Agents SDK (premium) - Building agents with OpenAI's official SDK
- Agent-to-Agent Protocol (premium) - Google's A2A protocol for agent interoperability
- Agent Guardrails & Safety (premium) - Input validation, output filtering, and safety boundaries
- Agent Observability (premium) - Tracing, logging, and debugging agent systems
- Agentic Workflow Patterns (premium) - Query routing, parallel execution, chaining, event-driven, tool orchestration
- Agent Deployment (premium) - Production deployment patterns and scaling strategies
- Durable Agents (premium) - Surviving crashes, deploys, and a human who answers on Thursday
- Full-Stack GenAI Application (premium) - Architecture walkthrough of production GenAI app (React + FastAPI + LangGraph + HITL)
- Autonomous AI Research (free) - By Andrej Karpathy - AI agents that autonomously run ML experiments overnight
- Project: Multi-Agent Research System (premium) - Production multi-agent research pipeline with LangGraph, Tavily Search, cost guardrails, and evaluation suite
- Project: Content Moderation System (premium) - Enterprise multi-agent content moderation with LangGraph, HITL, and guardrails
- Project: Due Diligence Analyst (premium) - 6-agent research system with parallel execution, fact-checking, contradiction resolution, and enterprise guardrails
- Project: Smart Claims Processor (free) - 7-agent insurance claims pipeline with LangGraph, CrewAI fraud crew, HITL checkpoints, agent memory, and country-aware config