AI Agents

Models that take actions in a loop instead of only answering.

AI Agents is taught here in 13 lessons, running from The agent loop through to Deploying and operating agents. Each lesson takes one topic, shows the working code, and links onward to the next, so the course can be read straight through in order.

Lessons

1. The agent loopHow an agent differs from a chatbot: a goal, a loop, tools, and a stopping condition — plus the failure modes of each.
2. Tools and function callingDescribing tools so the model can use them, validating arguments, and making writes safe.
3. Memory and retrievalContext windows, summarisation, retrieval-augmented generation and what to store outside the model.
4. Planning and task decompositionBreak a goal into steps a model can actually execute: plan-then-execute versus interleaved ReAct, replanning on failure, and when a fixed workflow wins.
5. Agent frameworks comparedWhat LangGraph, the OpenAI Agents SDK, CrewAI and AutoGen each abstract, the mental model behind each, and how to use one without marrying it.
6. Model Context Protocol in practiceWhat MCP standardises, how to run a server, exposing tools and resources, transport choices, and how to reuse one server across many clients.
7. Multi-agent patternsOrchestrator and worker, critic and reflection, context isolation and hand-off protocols — and why more agents rarely means more capability.
8. Human-in-the-loop and approval gatesInterrupt a run for approval, resume it safely, design escalation, and match autonomy to the risk of each action.
9. Guardrails, budgets and stopping conditionsStep and token ceilings, timeouts, cycle detection and spend caps — enforced in code, never requested in a prompt.
10. Evaluating agentsTask success rate, tool-call efficiency, recovery rate and cost per task, measured on a regression set you run for every version.
11. Observability and tracing agent runsStep-level traces of reasoning, tool calls and results, structured logs, and catching behavioural drift after a change.
12. Agent security and permissionsPrompt injection from tool output, least-privilege credentials, sandboxing, egress control, and auditing every side effect.
13. Deploying and operating agentsDurable state and resume, queues and concurrency limits, versioning prompts and tools, and a pilot-to-production rollout that does not surprise you.

More in AI & Intelligent Development

AI Basics Math for AI Machine Learning scikit-learn TensorFlow PyTorch Natural Language Processing LangChain Ollama OpenCV Codex Claude Code OpenCode Vibe Coding Selenium Playwright

FAQ

Last refreshed 2026-09-18.