AI Transformation Coach in London

You've rolled out AI somewhere real: the tools, the agents, the workflow changes that actually stuck.

So you already know the demo was never the hard part.

The hard part is getting a team to change how it works, and bringing along the people who aren't sure yet.

You'd coach working professionals to do in their own organisations what you've done in yours, and your job is to make them good at it.

What the job is

Your learners are working professionals in HR, operations, marketing and finance: the people who decide where AI belongs in their business and where it doesn't.

They stay in their jobs and work through a 15-month programme led by you, on real problems from their own organisation, from finding where AI fits to building agents for actual tasks, benchmarking tools, and making the business case with the return attached.

So you're not marking exercises.

You're reviewing real AI work going into real companies and coaching the judgement around it: framing the right problem, knowing when to trust a model and when not to, and winning over stakeholders who aren't convinced.

Business analysis applied to AI, done properly.

Your week

It comes down to one thing: your learners keep progressing at the right pace. Everything else serves that.

• Live coaching.

Each cohort has its own weekly two-hour session: a fixed slot, small group, scenarios and direct feedback rather than slides.

You'll run several across your caseload, and cover other coaches' sessions when they're on leave, so a typical week holds a handful of live slots.

• A caseload sized to your days: around 16-20 learners on two, 24-28 on three.

You know every one of them, and you catch the person who's quietly stalled early, because that's the part that decides whether they finish.

• One-to-ones and progression guidance.

Every learner gets a monthly one-to-one, plus information, advice and guidance (IAG) on progress, targets and next steps, individually and in groups.

• Quarterly reviews with learner and employer, a three-way conversation on qualification progress and the targets for what comes next.
• Marking and feedback.

Work marked within 48 hours, with feedback specific enough to act on, covering their real workplace AI projects and the portfolio evidence that counts towards the qualification.

• Onboarding and records. Suitability conversations with new joiners, and the record-keeping a regulated, inspected provider needs.
• Flexible by design

We built this as a real part-time role, not a full one cut down.

The fixed point is the live session, because the cohort needs a slot it can rely on.

You arrange everything else around your own week: marking, feedback, one-to-ones, keeping track of who's stuck.

What matters is that marking lands within 48 hours and every learner gets their monthly conversation, not whether you did it in one block or in half-hours between your own meetings.

Your caseload is one cohort, not a slice of somebody else's.

Why you'd want it

You'd be at the frontier all the time.

Your learners turn up with last week's tools and this morning's problem, across a dozen different businesses.

You'll stay sharper here than you would going deep in any one of them.

You'd get the thing corporate roles rarely give you: watching someone go from tentative to the person their whole organisation turns to on AI, and knowing you were the reason.

If that's the part of your current job you'd do all day if you could, this is the job.

And we're a startup, so how we coach isn't finished.

It's still being built, and we'd want you helping build it: how caseloads run, how we spot someone slipping sooner, how a new coach gets good faster.

There's real work here for someone who leans in rather than waiting to be handed the finished version.

Not as a route off coaching, though.

The people who improve how we coach are the ones deepest in it.

Who you are

You've worked hands-on with generative AI, agents or automation on real business problems, not experiments.

You can make a complex idea land with someone from a different background without a trace of condescension.

You stay current because you can't help it.

You're a startup person: you lean in, and you're comfortable where not everything's built yet.

And other people getting good at something gives you energy, because day to day, that's the job.

You don't need to have taught before.

We hire practitioners and teach them to coach, never the other way round: our coaches came from Accenture, Dialpad, Black Swan Data, BT and Three, and none of them arrived as teachers.

We fund a recognised teaching qualification (the Level 3 Award in Education and Training) and give you supervision, observation and honest feedback while you find your feet.

We're coaches, and that's the whole job

You have to want to develop people.

We do run some workshop-style teaching, and it's good, but the core is coaching: carrying a caseload and owning whether the people on it get there.

That's why a cohort keeps the same coach for all 15 months, since the continuity is most of why it works.

And it's why safeguarding, DBS and clean records matter here.

Not as background admin, but because someone's qualification depends on the evidence holding up.

• The essentials
• £40,000–£60,000 full-time equivalent, pro rata to your days, depending on experience.
• London hybrid (our office is in E1, by Liverpool Street), or remote elsewhere in the UK for the right person.
• Two or three days, fixed, agreed with you before you start.
• Enhanced DBS check, safeguarding and Prevent training, and the right to work in the UK; we can't offer sponsorship.

If you don't already hold a teaching or assessing qualification, you'll work towards one.

• Start as soon as you can.
• About i O-Sphere

We train people to do things with data and AI, not just learn about them.

Learners work on real problems inside their own organisations and in Prism, a simulated business we built on 500 million rows of real data. 900+ people have trained with us, we're rated 4.8 on Google, and we're one of the UK's fastest-growing data and AI training startups.

Teams at Pw C, BAE Systems, Dunelm, Motorway and CBRE send us their people.

How we hire

A short screening call, a conversation with the team you'd work with, a case study drawn from real delivery rather than an invented puzzle, and a final conversation.

We move quickly.

i O-Sphere is an equal-opportunities employer. We hire on merit and welcome applications from every background; if you'd need any adjustment to the process, just tell us.