Premium career track · Coming soon
Agentic AI Engineer Career Track
Design, build, test and ship AI agents that plan, use tools, remember and work together, safely.
- 13 modules
- 62+ lessons
- 2 modules free today
- 🎓 certificate
Get the early-bird price when it opens. Start the free modules today.
What you will be able to do
- Explain the plan → act → observe loop, and know when an agent is the wrong tool
- Give agents tools with function calling, JSON schemas, retries and error handling
- Add short-term, long-term and vector memory
- Build an MCP server and connect it to an agent
- Test agents with evals and tracing, and defend them against prompt injection
- Ship 4 capstone agents: research, support, coding and browser
Syllabus
-
1 Python Foundations ✅ Free now
Functions, dictionaries, JSON, files and errors: what every agent is built from.
Start the free course (20 lessons) ❯ -
2 Prompt Engineering ✅ Free now
Clear instructions, examples, roles and structured output: the language agents think in.
Start the free course (8 lessons) ❯ -
3 Agent Basics ⭐ Premium · Ready
The plan → act → observe loop, when not to use an agent, and your first agent with tools, errors and a step limit: 3 lessons with runnable Python.
- What an agent is: the plan → act → observe loop
- Workflows vs agents: when not to use an agent
- Your first agent in 40 lines of Python
-
4 Tool Use in Depth ⭐ Premium · Ready
JSON schemas, errors, retries, timeouts and idempotency, and designing tools an LLM uses well: 3 animated lessons with runnable, tested code.
- Function Calling and JSON Schemas in Depth
- Errors, Retries and Timeouts
- Designing Tools an LLM Can Use Well
-
5 Memory and State ⭐ Premium · Ready
The context window, long-term memory and summaries, and vector memory with embeddings: 3 animated lessons with runnable, tested code.
- Short-Term Memory: the Context Window
- Long-Term Memory and Summaries
- Vector Memory with Embeddings
-
6 Reasoning Patterns ⭐ Premium · Ready
ReAct, plan-and-execute and reflection with self-correction: 3 animated lessons with runnable, tested code.
- ReAct: Reason and Act
- Plan-and-Execute
- Reflection and Self-Correction
-
7 MCP: Model Context Protocol ⭐ Premium · Ready
How MCP works, building an MCP server, and connecting it to an agent: 3 animated lessons with runnable, tested code.
- What MCP Is and Why It Matters
- Build an MCP Server
- Connect an MCP Server to an Agent
-
8 Frameworks Compared ⭐ Premium · Ready
LangGraph, CrewAI, the Claude Agent SDK and OpenAI Agents SDK, and how to choose: 4 animated lessons with runnable, tested code.
- LangGraph in Depth
- CrewAI in Depth
- Claude Agent SDK and OpenAI Agents SDK
- Choosing (or Skipping) a Framework
-
9 Multi-Agent Systems ⭐ Premium · Ready
Orchestrator and workers, hand-offs and shared state: 2 animated lessons with runnable, tested code.
- Orchestrator and Workers
- Hand-offs and Shared State
-
10 Evals and Tracing ⭐ Premium · Ready
Test sets for agents, LLM-as-judge and tracing: 3 animated lessons with runnable, tested code.
- Test Sets for Agents
- LLM-as-Judge
- Tracing and Observability
-
11 Safety ⭐ Premium · Ready
Prompt injection in agents, permissions and human approval, and sandboxes: 3 animated lessons with runnable, tested code.
- Prompt Injection in Agents and How to Defend
- Permissions and Human Approval
- Sandboxes for Code-Running Agents
-
12 Production ⭐ Premium · Ready
Cost and latency, caching, deployment and monitoring: 3 animated lessons with runnable, tested code.
- Cost and Latency
- Caching
- Deployment and Monitoring
-
13 Capstone Projects ⭐ Premium · Ready
Research, customer support, coding and browser agents: 4 animated lessons with runnable, tested code.
- Capstone: Research Agent
- Capstone: Customer Support Agent with Tools
- Capstone: Coding Agent
- Capstone: Browser Agent