Case studies

Three stories of transformation — from a champions programme to a Knowledge Base living inside the client’s own systems.

Each case below is a different kind of work: a champions programme, a vibe-coding factory, a Knowledge Base the team queries while it writes. What stayed behind after we left: agents in production, internal champions, and rules written down.

Discuss a similar case →The Oper8 method →

Where we have rebuilt the work already.

Consumer health and pharma, under regulation and with data that cannot leave the perimeter. FMCG, with hard seasonal deadlines. Manufacturing and automotive, with a closed network. A startup ecosystem, with teams that ship weekly.

  • Coca-Cola HBC
  • Haleon
  • Bizerba
  • Chery
  • Astana Hub

Some engagements are under NDA, so the logo is not here and the story stays in the room. We are happy to talk through the closest match to your situation on a call.

Different problems, different formats, one Oper8 approach.

Each case is a different kind of work. Next to every title there is a line saying who it is a fit for, so you can tell straight away whether it speaks to your situation.

Champions programme

Haleon — 15 AI champions and the first agents in production within two months

A fit if

You need internal leaders who will start AI work in their own functions, not another course to sit through.

Brief

A large consumer health company wanted to get from basic literacy to working AI agents in a single programme, and to leave behind a community that keeps going on its own.

What we did

  1. 01

    Two-month intensive for 15 leaders

    Champions from marketing, finance, R&D and operations. Each brought a real problem from their own backlog.

  2. 02

    Weekly workshops

    Agent architecture, retrieval over internal documents, prompt design — worked through on the participants’ own cases.

  3. 03

    Mentoring all the way through

    Our team stayed alongside each champion, from the first spec to defending the project in front of the sponsor.

What the client kept

  • 15 AI champions with a mandate to keep starting initiatives in their departments.

  • The first agents running in production — document processing among them.

  • Projects defended in front of the sponsor at the end of the programme.

Format and duration:Eight-week intensive plus three months of mentoring.

Services:AI Champions 2.0 · Diagnostics

Vibe-coding factory

Astana Hub — one operating platform, assembled by the participants themselves

A fit if

Product managers, analysts and scrum masters want to build the tools they need without queueing for engineering.

Brief

Work was spread across Google Sheets, Notion pages and Slack integrations. The goal was a single operating system for scrum masters and project managers — built by the people who would use it.

What we did

  1. 01

    Seven vibe-coding modules

    No-code agent architecture, automation in n8n and Make, integrations with Slack, Notion and Asana. Every module ran on the participants’ own tasks.

  2. 02

    Mentoring from first automation to production

    We stayed with the teams until an agent was actually running inside their process, not until the module was over.

What the client kept

  • A single operating platform, assembled by the participants rather than handed to them.

  • Scrum masters and project managers work in one place instead of a spread of sheets and integrations.

  • Participants report saving around ten hours a week on routine work.

Format and duration:Seven intensive modules plus mentoring — six to eight weeks in total.

Services:Vibe-coding factory · AI transformation

Knowledge Base in the client’s own environment

US professional services firm — a decade of proposal material the team can finally query

A fit if

Years of proposals, reports and templates sit across shared drives, and finding the right precedent takes a day.

Brief

A consultancy bidding for public-sector work had ten years of written material and no way to reach into it. The task was to make that archive answerable while a bid is being written, not searchable after the fact.

What we did

  1. 01

    Knowledge Base inside the client’s environment

    The archive stays in the client’s own systems. Nothing is uploaded to an outside service.

  2. 02

    Agent skills on top of the base

    The team queries the base from the tools it already writes in, so looking something up is part of the work rather than a detour.

  3. 03

    A weekly audit that keeps the base honest

    An agent reviews what changed and flags material that has gone stale, so the base does not quietly rot.

What the client kept

  • The Knowledge Base became the team’s primary workspace, not a second place to look.

  • Agent skills sit on top of it, so a decade of precedent is available mid-sentence.

  • The weekly audit keeps the content current without anyone owning that as a chore.

Format and duration:The client is not named: we do not have written consent to publish it. No ROI figures are quoted — the numbers in the project material are modelled, not measured.

Services:AI transformation

Different problems, the same way of working.

In none of these engagements were we selling a course on AI. We came in through diagnostics — to find out what was actually in the way — and left through an artefact the client keeps: a training track, a community of champions, a platform, a set of agents.

In our experience that is the difference between a transformation that holds and training that is forgotten by the next quarter.

  1. 01 · Step

    We come in through diagnostics

    We look at what is actually in the way. Not selling a course on AI — first working out where the flywheel has jammed.

  2. 02 · Step

    We leave through an artefact

    Something that works stays with the client: a training track, a community of champions, a platform, a set of agents in production.

  3. 03 · Step

    We leave a mandate and a memory

    Internal leaders with permission to act, a log of decisions, and the metrics behind them — that is what separates a transformation from training everyone forgets by the next quarter.

More on the Oper8 method →

Thirty minutes to look at your situation and say which of the four formats fits.

Tell us your industry, the size of the company and the problem you are chasing. We will point to the case closest to yours and say which step is worth taking first.

01 · Diagnostics

If it is not clear where to start.

02 · Training

If you need literacy at scale, or a champions programme.

03 · Platform

If the data cannot leave your perimeter, we build inside it.

04 · Everything

If it is a full transformation: diagnostics, platform and people.

Book a 30-minute call →Telegram

Related reading

The Oper8 method

The open method: the Data Flywheel, five elements of the flywheel, five principles.

Read the method

AI potential diagnostics

The first step: a maturity map, a process audit, an honest read on readiness.

Book a diagnostic

Transformation services

Pilot, process redesign, fractional CDO, an internal AI centre of excellence.

See the services

Let’s
work together

We’ll look at your problem, pick the right format, and show where AI actually strengthens your operations.

Book a consultationTelegram

50+

companies have gone through AI transformation with us — from diagnostics to working solutions.

Kiberry

© 2026 Kiberry — the European practice of ScrumTrek

Privacy policy