Functional AI: Use AI
No coding. Eight modules take you from choosing the right AI tool to deploying agents, RAG systems and automated workflows, with business examples from many industries.
Foundation now. Architect next.
Functional AI comes in two levels. Foundation is open for enrolment today. Architect goes further and is still being built.
Foundation level
Use and operate AI at work, with results you can verify.
- No coding required
- Free Module 0, then 8 modules and 32 sections, about 56 to 63 hours including Module 0: roughly 4 weeks at 14 hours a week, or 7 to 8 weeks at 8
- Tools, prompting, automation, APIs, RAG, agents, MCP, data and deployment
- You build a tested skill, a running agent and a two-agent workflow
- Pass 80% on each quiz to unlock the next section
- Lifetime access, one payment
Already enrolled? Log in
Architect level
Design, evaluate and govern AI solutions, not just use them.
- Everything in Foundation
- A Deep Dive that rebuilds each module's system in code: APIs, RAG, MCP servers and pipelines
- A capstone project
- A 1:1 session with a senior AI architect before the capstone
- An Architect certificate
No date or price yet. Foundation students can upgrade by paying the difference.
Who it's for
- Working professionals in any function: marketing, finance, HR, operations, sales
- Founders and small-business owners who want AI inside their business
- Accountants, lawyers, consultants and other practitioners
- Anyone who has tried ChatGPT and wants to go much further, without becoming an engineer
What you'll be able to do
- Choose the right AI tool for each job and evaluate new ones
- Write prompts and reusable templates that give consistent business-grade output
- Automate multi-step workflows with no-code tools
- Connect AI to apps through APIs, safely
- Build a RAG assistant that answers from your own documents
- Build a tested skill, a running agent and a two-agent workflow
- Process documents and data (invoices, reports) with AI and visualise results
- Deploy, monitor and control the cost of AI in production
8 modules, 32 sections
MODULE 1AI Tools Ecosystem
Map the whole AI landscape and build your personal toolkit.
- The AI landscape
- Chat assistants in depth
- Specialised tools
- Evaluating and choosing tools
MODULE 2Prompt Engineering Mastery
Communicate with AI so it gives precise, useful results, from basics to business patterns.
- Prompt fundamentals
- Advanced techniques
- Patterns and templates
- Business applications
MODULE 3Workflow Automation with AI
Automate repetitive work with no-code tools and multi-step AI workflows.
- Intro to workflows
- No-code tools
- Multi-step workflows
- Optimisation
MODULE 4APIs and Integration
Connect AI to real applications through APIs.
- API fundamentals
- AI APIs
- Building apps
- Security practices
MODULE 5RAG and Knowledge Systems
Build AI that answers from your own documents and data.
- What is RAG
- Knowledge bases
- Vector databases
- Production RAG
MODULE 6AI Agents, Skills and MCP
Go from assistant to autonomous worker: a tested skill, a running agent and a two-agent workflow.
- Understanding agents
- Building a skill (tested)
- Building an agent (running, with a cost cap)
- Two-agent workflow (content pipeline)
MODULE 7Data Processing with AI
Turn spreadsheets, documents and databases into insight without a data-science degree.
- Data analysis
- Document processing
- Data pipelines
- Visualisation
MODULE 8Deployment and Scaling
Take AI solutions from prototype to production for real users.
- Deployment
- Monitoring
- Cost optimisation
- Scaling (graduation)
Learn
Each section opens with its learning objectives and where the skill shows up in a 2026 job, then 12 or more teaching cards you open and study.
Try
"Try it yourself" prompts you paste into Claude or ChatGPT.
Prove
A 10-question quiz per section. Pass at 80% to unlock the next.
Things that work, not just notes
- A tested skill you can drop into your own work
- A running agent that carries out a multi-step task, with a cost cap and a run log
- A two-agent workflow (a content pipeline) with a person reviewing the result
- A knowledge assistant that answers from your own documents, with an evaluation
The quizzes check your judgment on realistic scenarios. The builds are where you show the skill: each one has a written checklist you test your result against, and you keep the evidence.
Sample build task
From the agents module: build an agent that looks up order status in a read-only spreadsheet and saves a draft reply for a person to approve. You write its spec first (goal, tools, permissions, approval points, stop conditions, cost cap), run seven tests including a prompt-injection attack planted in the data, read the execution log, and measure real usage. You hand in the spec and the test log, and they are scored against a published rubric.
Sample quiz question
Which design best follows least privilege for an order-status agent?
Correct answer: a read-only lookup on one tab and a write tool limited to a drafts tab, with no email tool.
Why: each tool should be the narrowest one that does the job. Sending, editing and web access add risk the task does not need.
Match the modules to your goal
Business owners and professionals
Modules 1, 2, 3, 6 and 7 map most directly to improving a business: tool choice, prompt templates, no-code automation, agents and skills, and processing invoices, reports and spreadsheets. Start here.
Concepts first
Modules 4, 5 and 8 (APIs and security, RAG and vector databases, deployment and monitoring) are taught without code, but they are the most technical. Expect to understand them and direct others; building them alone is easier if you are comfortable with technical tools. To build these in code, see Core AI or the planned Architect level.
This is a tools-and-building-blocks program, not a business-strategy course. Choosing your highest-value use cases is something you apply with it.
Ready to start?
Not sure which track? The free 100-question assessment tells you.
