No Hype AI

Does your AI actually work?


Hands-on AI lessons and coaching — from the basics of prompting and context management, through agents and orchestration, to running your own local models. Vendor-neutral, and taught by a senior practitioner.

No hype, just experience from 13 years of hands-on AI and 30 years building and delivering software.

First lesson

September 2026

What the lessons cover

Ten courses, 140 lessons. Written; being recorded now. Each one is a job you can do at the end of it.

  1. 01 Evaluation & reliability Build an eval set, measure what your system actually gets right, and tell whether a change made it better or worse.
  2. 02 Prompting & context Get output you can repeat, rather than a good first attempt you cannot reproduce.
  3. 03 Building with AI Do real work in Claude, Cursor and Codex — including the parts where the tool is wrong and you have to notice.
  4. 04 Agents & orchestration Build an agent that finishes a job end to end, then run several of them without losing track of what they did.
  5. 05 RAG & knowledge systems Put your own documents behind a model, and check the answers really came from them.
  6. 06 Local LLMs Run a model on your own hardware, and work out when that is the right call and when it is not.
  7. 07 AI product judgement Decide what should use AI at all, and what a user will trust it to do.

The email list

Channels

The lessons go out on YouTube. The handle is the same everywhere — @no.hype.ai.expert — except X, which is @nohypeaiexpert. LinkedIn is Simon’s personal account.