AI mentoring & education

Helping young builders take an idea all the way to the real world, safely.

I mentor young people who are curious about AI, technology and starting things of their own. Learning how AI works is the starting point. The lessons that stay with them come later: putting something in front of real users, looking after those users, and learning from what happens next. I bring three years of teaching children in Japanese schools, twenty years of taking products to real users, and a background in privacy and user safety at Google and Bumble.

See a sample programme Get in touch
Portrait of Rory McClelland
Rory McClelland, London
3 yrsteaching in Japanese elementary and junior high schools
20 yrstaking products from idea to real users
2privacy and safety leadership roles: Google Chrome, Bumble
MScComputer Science, Distinction, Birmingham
01 · Why me

Three things a young builder rarely finds in one mentor.

A teacher first

I know how children of this age learn

I taught for three years in elementary and junior high schools in Shiga, Japan, including special needs classes. That is where I learned what keeps children of this age motivated, what makes them switch off, and how to explain a difficult idea without talking down to anyone.

Someone who ships

Real users, beyond the demo

For twenty years my job has been getting products out of the building and into people's hands: a new-to-market app at Soho House, privacy features in Google Chrome, and AI pilots running on real data. The same discipline is taught here, at a smaller scale.

Safety built in

Privacy and safety from the inside

I led privacy for Google Chrome and product safety at Bumble. A young builder learns to protect their own data and their users' data, and to ask who could be harmed by what they make before it launches, instead of afterwards.

02 · The gap

Most AI tuition stops at the demo.

Building a working demo with an AI tool is now an afternoon's work. Understanding how AI works, and using the tools well, is covered by most courses. The harder and more valuable lessons start afterwards: when real people use what you have made, when something breaks, and when you are responsible for other people's data.

1

Understand

How AI works, in plain terms.

2

Use

Getting good results from AI tools.

3

Build

A working prototype.

4

Ship

Real users, real feedback, a launch.

5

Run safely

Privacy, safety, looking after users.

Typical AI tuition
This programme

The last two rungs are where product development, marketing, project management and responsibility are actually learned, because they stop being theory. They are also the rungs I have spent my career on.

03 · A sample programme

Twelve weeks, from first idea to something real.

The programme follows the same five-stage method I use with businesses, sized for a young builder of roughly 11 to 16. Each stage ends with something tangible, and each one adds a safety habit. It is shaped around the student's own interests, so choose a track to see how it changes.

What is learned

What gets made

Safety habit

In practice: a game

A sample only. The pace, projects and balance between building, marketing and business are agreed with the student and family at the start, and revisited at the end of each stage.

04 · Safety & governance

Governance, sized for a young builder.

With businesses, I spend a good deal of time on AI governance: who can see the data, what a system is allowed to do, and who is accountable when something goes wrong. The same questions apply to a young person's first app, in simpler words. Most young builders are never asked them, and they are far easier to learn early than to unlearn later.

My data

What should never be typed into an AI tool, how to check a tool's privacy settings, and why 'free' usually means the user is paying some other way.

Tools and ages

Many AI tools set minimum ages in their terms of use. Tools and accounts are chosen and set up with parents, and parents can see what is being used and why.

My users' data

Collect the minimum, keep it safe and delete it when it is no longer needed. In the UK, services likely to be used by children must follow the ICO's Children's Code, which is a useful checklist in its own right.

Checking the machine

AI tools can be confidently wrong, and can be steered by text hidden in what they read. So their work is checked, and nothing important is left to them unsupervised.

Before you launch: a quick check

A simplified version of the risk check I use with businesses. Tick what applies to the project.

Ready for testers

    05 · How it works

    Working with families.

    Sessions

    Mostly online, and scheduled around the school timetable and term dates. Occasional in-person sessions are arranged where they help, for a launch or a demo day, for instance.

    Keeping parents informed

    A short written update at the end of each stage: what was learned, what was made, and what comes next. Parents are invited to the end-of-programme demo.

    Independence

    The aim is a student who can research, plan and solve problems on their own. Support is stepped back as confidence grows, which is also how autonomy is handed to AI systems in business.

    Checks

    I am happy to undergo enhanced background checks, and to agree ground rules for online sessions with the family at the outset.

    06 · Contact

    Get in touch about mentoring.

    Email is the best way to reach me. I am happy to talk through what a programme could look like for a particular student, and I read and reply to email myself, usually within a day.

    Emailhello@mcclelland.info LinkedInlinkedin.com/in/rorymcclelland Main siteProduct & AI transformation