Executive briefing (90 min)
What AI can and cannot do for your organisation, where the money is, what the real risks are, and what questions to ask a vendor. Aimed at boards, C-suite and senior managers. No hype, no product pitch.
Service · Enablement
AI training your team will actually use on Monday
Most corporate AI training is a slideshow about the future of work. Ours is a working session where your staff use AI on their own real tasks, save what works, and learn exactly which information must never leave the building.
Executive briefingsHands-on workshopsDepartment tracksPrompt librariesAI policy
What is corporate AI training?
Corporate AI training is structured, practical instruction that enables an organisation’s staff to use artificial intelligence tools effectively and safely in their actual jobs. Good programmes are role-specific and hands-on: participants work on their own real tasks, learn how to evaluate and correct AI output rather than trust it, and leave with a clear understanding of what data may and may not be entered into which tools.
There is an uncomfortable fact behind most training requests: your staff are already using these tools. They are using them on personal accounts, with company documents, with no policy and no oversight. The training question is therefore not “should we adopt AI” but “how quickly can we make what is already happening safe and productive?”
The productivity upside is real but uneven. Some staff will save several hours a week almost immediately; others will find little use in their role and should not be pushed. We are honest about that in the room, because overselling is what makes people quietly stop.
Signs you need this
- Staff are using AI tools on personal accounts with company data.
- You have no AI policy, or one nobody has read.
- Two people are getting great results and nobody else knows how.
- A team is producing AI text that is obviously AI text.
- Your board wants to understand AI without a vendor pitch.
- You bought licences and adoption stalled after two weeks.
In practice
The sessions we run
Delivered on-site anywhere in Kenya or online. Most clients start with the executive briefing and one departmental workshop.
Hands-on staff workshop (1 day)
Up to 20 people, laptops open, working on their own real tasks. Ends with each person having a saved set of prompts that work for their specific job.
Department tracks
Separate content for finance, sales, HR, operations, marketing and customer service — because a useful example for a credit controller is useless to a marketer.
Safe & compliant use
What may be pasted where, how to spot a confident wrong answer, how to handle personal data under the Data Protection Act, 2019, and how to document AI use in regulated work.
Technical track (2 days)
For developers: APIs, structured outputs, tool calling, retrieval design, evaluation, cost control, and the production failure modes nobody warns you about.
Champions & follow-up
A 30-day clinic to unstick people, plus coaching for internal champions so the capability stays in your organisation after we leave.
The same answer, two ways
What a training day actually looks like
People bring their real work, and leave having done it faster.
We start with twenty minutes of honesty: what these tools genuinely do well, what they get confidently wrong, and why the difference matters when you are the one who signs off the output. No science fiction, no slide about robots taking jobs.
Then everyone opens their laptop and picks something real — the weekly report, the customer email backlog, the tender response, the job description they have been avoiding. We work through it together. This is where the room usually goes quiet and then gets loud, because seeing your own task take four minutes instead of forty is more persuasive than any statistic.
We spend a solid block on checking, because the biggest risk is not that AI refuses to help, it is that it helps confidently and wrongly. People learn to spot the specific ways these tools go wrong: invented figures, plausible but false citations, and quiet confidence about things they cannot know.
We finish with the rules: what may be typed into which tool, what must never be, and who to ask when unsure. Everyone leaves with a written prompt library for their role and a one-page policy summary that fits on a noticeboard.
How we measure it
Not by a feedback form. Thirty days later we come back and ask what people are actually still doing. That number is usually between 40% and 70%, and it tells you far more than a smiley face on an evaluation sheet.
Role-based curriculum, task-anchored practice, and an explicit data-classification model rather than a vague “be careful”.
Needs analysis. Before the session we sample real artefacts from each department — reports, emails, spec documents, spreadsheets — and build exercises from them. Generic exercises produce generic engagement; people disengage the moment an example is not their job.
Capability model. We teach a working mental model of what current systems are: next-token predictors with strong pattern generalisation, no reliable internal fact store, no native arithmetic guarantee, and a context window that behaves like working memory. This single framing explains most observed failure modes and is more durable than a list of tips that expire with the next model release.
Prompt practice. Role and task framing, few-shot examples drawn from the participant’s own past work, explicit output structure, decomposition of multi-step tasks, and iterative refinement. We teach verification patterns explicitly: asking for sources, asking for the reasoning, requesting the counter-argument, and cross-checking numbers outside the tool.
Data classification. A four-tier scheme mapped to your actual document types — public, internal, confidential, restricted — with a concrete tool matrix showing what may be entered where. This is the part that satisfies your Data Protection Act, 2019 obligations, and it is the part most training programmes skip because it requires knowing your business.
Technical track. For engineers: token economics and cost modelling, structured output enforcement, tool and function calling, retrieval architecture and chunking trade-offs, evaluation harness design, prompt versioning, latency and streaming, rate limiting and graceful degradation, and provider abstraction so a deprecation is survivable.
Reinforcement. Curated prompt library in your own wiki or drive, a 30-day clinic, and internal champion coaching with an agreed cadence.
What you get
What your organisation keeps
Training that leaves artefacts behind, not just a good afternoon.
- A prompt library written for your roles, your documents and your terminology.
- A one-page data classification and tool matrix for the noticeboard.
- A draft AI acceptable-use policy, ready for HR and legal review.
- Session recordings and slides for staff who could not attend.
- A 30-day follow-up clinic to unstick people.
- Internal champion coaching so the capability stays with you.
- An honest report on where AI helped and where it did not.
Typical engagement
TimelineBooked 2–3 weeks out; delivered in a day
Indicative investmentFrom KES 120,000 per training day (up to 20 people)
Who we work withBoards, managers, frontline staff and developers
You ownCode, prompts, docs and accounts — outright
Every quote is fixed before work starts. If we scoped it wrong, that is our risk, not yours. See how pricing works.
Typical stack
On-site anywhere in KenyaOnline deliveryEnglish & KiswahiliUp to 20 per sessionSector-specific contentCertificates of attendance
We choose tools your team can maintain, not the ones that make us look clever.
Questions
AI Training for Teams — your questions answered
What does AI training for a team actually involve?
A working session, not a lecture. People bring laptops and their own real tasks — a report they write every month, a proposal they are drafting, a spreadsheet they wrestle with — and leave having done that task faster with AI, with a saved set of prompts that work for their job. We cover what these tools are good and bad at, how to get useful output, how to check it, and crucially what must never be pasted into a public tool.
How much does AI training cost in Kenya?
A full training day for up to 20 people typically starts around KES 120,000, whether on-site in Nairobi or online. Executive briefings of 90 minutes cost less; multi-department programmes with follow-up clinics cost more. Travel outside Nairobi is charged at cost. We would rather quote for what you need than sell a fixed package.
Our staff are already using ChatGPT. Is that a problem?
It is a risk you should size quickly rather than panic about. The real exposure is confidential information — customer lists, salary data, draft contracts, patient records — being pasted into consumer tools, which may be a breach of your obligations under the Kenya Data Protection Act, 2019. The answer is rarely a ban, because bans move the behaviour underground. It is a clear policy about what may go where, an approved tool people actually prefer, and one training session so everyone knows the line.
Do you train non-technical staff?
That is most of our work. We run separate tracks for executives, managers, and frontline staff in finance, sales, HR, operations and customer service, because the useful examples are completely different for each. No coding is required in any of these sessions, and we never assume prior knowledge.
Can you train our developers as well?
Yes — a separate technical track covering the API surface, structured outputs, tool calling, retrieval design, evaluation, cost control and the failure modes that only show up in production. This is usually a two-day workshop built around a system your team is actually trying to ship.
What happens after the training day?
Skills decay fast without reinforcement, so a single day rarely sticks on its own. We include a prompt library tailored to your organisation, a 30-day follow-up clinic to work through what people got stuck on, and optional internal champion coaching so the capability lives inside your team rather than leaving with us.
Keep exploring
Other ways we help
Most clients combine two or three of these.
AI Workflow Automation
We find the repetitive work eating your team's week — reconciliation, data entry, approvals, reporting — and hand it to software that never gets tired.
Explore AI Workflow Automation
AI Agents & Chatbots
A WhatsApp and website assistant that answers customers in English and Kiswahili, qualifies leads, checks order status and escalates to a human when it should.
Explore AI Agents & Chatbots
AI Strategy Consulting
An honest map of where AI will make you money, where it will waste your money, and what to do first — costed, sequenced and tied to your numbers.
Explore AI Strategy Consulting