Advanced · module A10
AI Agents and Autonomous Systems
By the end: Build reliable single-agent and multi-agent systems with tools, memory, guardrails and evaluation
- Covers
- LLM agents that plan, use tools, remember and work together, built safely
- Contents
- 2 chapters, 11 topics
- Time
- About 3 hours
- Level
- Advanced
- Your progress
- 0 / 11
This module builds on Retrieval-Augmented Generation and Knowledge Systems. We suggest looking at it first, but the choice is yours.
Chapter 1 Agent Foundations
- 1.110 minBeing written
What Is an Agent?
Agents compared with workflows, and when not to use one
Decision exercise
- 1.215 minBeing written
Tool Calling and Function Schemas
Describing tools, executing calls and handling errors
Code: an agent with three tools
- 1.315 minBeing written
The Model Context Protocol
Connecting agents to tools and data in a standard way
Code: build a small MCP server
- 1.415 minBeing written
Planning and Reasoning Patterns
ReAct, plan-and-execute and reflection
Code: a planning agent
- 1.512 minBeing written
Agent Memory
Short-term, long-term, episodic and semantic memory
Code: add memory to an agent
Chapter 2 Advanced Agents
- 2.118 minBeing written
Multi-Agent Systems
Orchestrators, handoffs and debate
Code: a research team of agents
- 2.215 minBeing written
Computer-Use and Coding Agents
Agents that operate browsers, terminals and codebases
Lab: a sandboxed coding agent
- 2.312 minBeing written
Evaluating Agents
Task success, trajectories and regression tests
Code: an agent test suite
- 2.412 minBeing written
Guardrails, Permissions and Human Approval
Least privilege, approvals and audit logs
Lab: add guardrails
- 2.545 minBeing written
Project: An Autonomous Research Assistant
Plan, search, read, write and cite
Project
- 2.620 minBeing written
Module Review and Assessment
Coding tasks and a scored quiz
Graded project