Subject
Agentic AI
Build programs that use language models, tools and memory to do real work, from first call to production.
Foundations
Engineering habits, how models work, and how to call them reliably.
Level 0
Engineering foundations
Build, test and ship ordinary software before adding models. Every later level assumes these habits.
8 modules · about 4 weeks
Level 1
How LLMs work, for builders
Build a working mental model of tokens, context and sampling, so model behaviour stops being surprising.
8 modules · about 3 weeks
Level 2
LLM APIs and structured output
Turn single model calls into dependable software: streaming, validated output and the first tool call.
8 modules · about 3 weeks
Level 3
Prompting and workflows
Write prompts that behave the same way tomorrow, and learn the five workflow shapes that solve most problems without an agent.
7 modules · about 3 weeks
Knowledge
Give a model the facts it was never trained on.
Level 4
Retrieval and RAG for agents
Ground a model in your own documents, measure how well retrieval works, then let an agent decide what to look up.
10 modules · about 5 weeks
Agent core
Loops, tools, memory and teams of agents. The heart of the path.
Level 5
Your first agent
Build the observe, decide and act loop yourself on the raw API, with limits, traces and a way to stop.
8 modules · about 4 weeks
Level 6
Tools, MCP and computer use
Connect agents to real systems through standard protocols, and keep those connections safe.
9 modules · about 4 weeks
Level 7
Memory and context engineering
Decide what the model sees at each step, what it remembers across sessions and what lives outside the window.
8 modules · about 3 weeks
Level 8
Multi-agent systems and frameworks
Learn when several agents help and when they only add cost, then compare the main frameworks on your own workload.
8 modules · about 4 weeks
Trust
Prove it works, then make it hard to misuse.
Level 9
Evaluation and observability
Test systems that never give the same answer twice, and see what they do once real users arrive.
10 modules · about 4 weeks
Level 10
Safety, security and governance
Assume text from the outside world is hostile and design the agent so one bad input cannot do real damage.
10 modules · about 3 weeks
Production
Run agents for real users without losing sleep.
Level 11
Production architecture and system design
Run agents as services: queues, durable execution, streaming, cost control and the design reasoning interviewers test.
11 modules · about 5 weeks
Level 12
Agent products and patterns
See how research, coding and support agents are built in practice, and design the interface people will trust.
6 modules · about 3 weeks
Mastery
Go deeper, ship a public capstone and prepare for interviews.
Level 13
Advanced topics
Go deeper where the evidence says it pays: tuning small models, running them efficiently and reading research.
9 modules · about 4 weeks
Level 14
Capstone and career
Ship one public, production-grade agent and practise explaining every decision behind it.
6 modules · about 4 weeks