Development & Programming
In-Person
Online
beginner Level
AI Automation
Starts 12 October 2026•6 months
Overview
About this programme
The complete AI Automation programme, and deliberately a no-code one. It continues straight on from AI Automation Foundations into advanced workflows, business-system integration, guardrails and running automations in production. Anyone who wants to build automations in code should take AI Engineering: The Developer’s Path instead.
Curriculum
Programme structure
Prerequisites
None, and no programming is required at any point. The first six modules are shared with AI Automation Foundations, so Foundations graduates continue from module seven.
What you'll learn
- Automate real business processes end to end, with no programming
- Build advanced workflows in n8n and Make, including multi-step agents
- Connect the systems a business actually runs on — CRM, ERP, email, sheets
- Ground automations in a company’s own documents and knowledge
- Put evaluation, guardrails and human review around AI decisions
- Handle security, privacy and compliance for automated work
- Run automations in production: cost, monitoring and reliability
12 modules
- What today's AI can and cannot be trusted with, and how to tell the difference before you build. Map a process you do by hand, find the parts worth automating, and set the standard the finished automation has to meet.
- What a large language model actually does, in plain terms
- Where AI is reliable, where it is not, and how to tell
- Mapping a manual process end to end
- Choosing what to automate first: frequency, cost, risk
- Defining "done" — what good output looks like
- The tool landscape: assistants, workflows, agents
- Cost, data and privacy considerations before you start
You'll build: a map of a manual process, scored for what is worth automating first
Module 1: How AI Automation Works
What today's AI can and cannot be trusted with, and how to tell the difference before you build. Map a process you do by hand, find the parts worth automating, and set the standard the finished automation has to meet.
- What a large language model actually does, in plain terms
- Where AI is reliable, where it is not, and how to tell
- Mapping a manual process end to end
- Choosing what to automate first: frequency, cost, risk
- Defining "done" — what good output looks like
- The tool landscape: assistants, workflows, agents
- Cost, data and privacy considerations before you start
You'll build: a map of a manual process, scored for what is worth automating first