This course contains the use of artificial intelligence.

Most analysts have tried AI, received something that read well and meant little, and quietly gone back to writing everything themselves.

The output was mediocre because the input was a sentence. What produces usable work is a structure, and the structure is learnable.

Why the first attempt disappoints

You asked for a document and got a plausible one, with the specifics invented. You asked for analysis and got a summary of what you had already written. The model had no access to your policies, your definitions or your templates, so it answered from the general internet. And every time you wanted the same output again, you rebuilt the request from scratch, which means nothing accumulated.

None of that is a limitation of the technology. It is what happens when a tool designed to run on context is used without any.

What the course covers

Thirty-five lessons in three layers. The analyst's craft first, twelve lessons: framing a problem before writing a requirement, user stories with acceptance criteria, AS-IS and TO-BE modelling, impact analysis, and describing processes in BPMN — because a process nobody has described cannot be automated, only accelerated in its confusion.

Then the AI fundamentals: what these models can and cannot do, working with files, building a custom assistant on your own documents. Then prompt engineering properly — the seven blocks of a prompt, the four classic frameworks, chain-of-thought and tree-of-thoughts, ReAct, and the shift from prompting to context engineering with Projects, RAG and memory.

Building things that keep working

Then eight lessons on the operations layer, which is where this stops being about chatting. Projects with their own instructions and knowledge base, custom Skills written as plain text files so a workflow runs identically every time, connectors to your drive, mail and messaging through MCP, retrieval over your own documents, and packaging the whole set into a plugin your team installs with one command.

And finally five lessons of analysis end to end: a raw system export turned into a clean dataset, descriptive analysis without manual formulas, diagnosis of causes and anomalies, forecasting through interpretable rules rather than an unexplainable model, and a board-ready presentation defended in front of people who will push back.

A note on the examples

The AI sections are taught on people and operations data — recruiting, onboarding, reviews, pay. The methods are domain-neutral: a Skill is a Skill, a RAG setup is a RAG setup, and the PEARL analysis cycle works on any dataset. I have kept the worked examples as they were recorded rather than genericising them, because a real dataset teaches more than an invented one.

Who is teaching this

Mike, the number one HR instructor on Udemy. More than 1.6 million course enrolments, over 150,000 professionals trained, PHRi and SHRM-CP certified, HRCI representative in more than 10 countries. I built the people function of the unicorn Preply and worked at Wargaming, Alfa-Bank and iDeals.

What is included

• Lifetime access to all course materials

• Active instructor support in the Q&A section

• Udemy Certificate of Completion

• Practical assignments and real business cases

• A section with additional courses, tools and resources

Take the document you write most often and build it as a Skill during the course. That single artefact usually pays back the time before you finish. Enrol now and start the first lesson today.