← Back to Dashboard
Module 06 · Days 36–42
AI Agents & Agentic Systems

The fastest-growing area in applied AI 2025. Agents plan, use tools, browse the web, write and run code, and collaborate. By end of this module you ship an autonomous research agent.

D36
Agent fundamentals: ReAct, tool calling & planning loop
The reason and act loop, function calling via APIs, and the mental model every AI engineer needs.
D37
Agent memory: short-term, long-term & episodic
In-context summarisation, vector memory, structured key-value storage. How agents remember across sessions.
D38
Multi-agent systems: CrewAI, LangGraph & orchestration
Supervisor and worker agents, message passing, parallel execution. Build a 3-agent pipeline.
D39
Code execution agents & browser automation
Agents that write and run code, browse the web, fill forms, scrape data. Playwright plus sandbox.
D40
Agent evaluation, reliability & guardrails
Testing agents systematically. Task success rate, cost per run, hallucination in tool calls, content guardrails.
D41
Production agents: caching, cost control & observability
Semantic caching, model routing, LangSmith tracing, and session cost budgets for production.
🏗
🏗️ PROJECT: Autonomous Research & Report Agent
Agent takes a question, searches the web, reads PDFs, runs data analysis code, and delivers a structured report.