Innovator — AI Fluency
Eight weeks that take you from building agents that work to building agents that hold up.
Builder gets an agent running. Innovator is where you find out why most agents fall over in their second week — and how to build the kind that does not.
The first four weeks are the complete Builder foundation. The next four go under the hood: how an agent is actually assembled, what happens to your documents long before a model sees them, why retrieval returns confident nonsense and how to trace the cause, and how to connect an agent to real systems through APIs, tools and MCP so it can act rather than only answer. You give it memory, so it stops starting cold every time.
Two projects across sixteen live sessions, one at the end of each stage, and you keep both.
This is also where the work becomes certifiable: Innovator qualifies you for the AI Fluency exam, and for the portfolio internship that follows it.
Sessions are taught live on fixed dates, two a week, in a small batch, and your group starts on the date shown at checkout. Covered by a 15-day refund — no questions asked.
What you will learn
- Take an agent apart and diagnose why it misbehaves
- Ground a system in your own material with retrieval that you can trust
- Connect agents to real systems through APIs, tools and MCP
- Give an agent memory that stays useful as it grows
- Finish two working AI projects you keep
- Qualify for the AI Fluency exam and certification
Syllabus
Month 1 · project: First Line Agent
Foundations, and your first agent
- 1.AI Introduction
- 2.Basics of LLM
- 3.AI Prompt Engineering
- 4.AI Hygiene
- 5.Automation, AI agents and agentic work
- 6.Action Planning
- 7.First Line Agent — project, day 1
- 8.First Line Agent — project, day 2
You ship a working agent, from a standing start.
Month 2 · project: Agentic Workflow
How agents actually work
- 1.Anatomy of Agents
- 2.LLM Preprocessing
- 3.Vector Databases
- 4.RAG Fundamentals
- 5.API Wrappers, Tools, MCP
- 6.Agent Memory
- 7.Agentic Workflow — project, day 1
- 8.Agentic Workflow — project, day 2
You build an agent that remembers, retrieves and uses real tools.