Training

Generative AI in practice

Ask the same question a different way and the answer can be far more useful. This course teaches how to ask, and how to turn a set of prompts into a reusable assistant (a Gem).

Who it’s for

Staff use AI, but the answers are often generic and need heavy editing
Everyone is figuring it out alone, and the company has no shared way of using it
You want AI to help with analysis and decisions, not just write copy
You have a lot of internal documents and want AI to answer based on them

Engagement cycle 3 months, 6 months or a year
Pricing Estimated per engagement

Email us

A question

Something your work really needs solved

Structured prompt

Role, task, format

AI-assisted reasoning

Compare options, spot blind spots

Custom assistant

Saved as a reusable Gem

For the same question, how you ask decides whether the answer is usable; a prompt that works well can be saved and reused.

What we do

Structured prompts

Writing clear instructions with four elements, role, task, context and format, so the output is predictable.

Reasoning and decisions

Guiding AI to reason step by step (chain of thought) to compare options and spot blind spots, not just generate content.

Knowledge base

Using NotebookLM to turn company documents into a knowledge base you can talk to, with sources cited.

Custom assistants (Gems)

Turning a department’s common prompts into a reusable Gemini assistant, so the whole team works the same way.

Usage policy

Which data can be given to AI, and how to check its output.

How it works

1. Needs interview

Learn what the department does and how staff use AI today.

2. Course planning

Design exercises around your actual work.

3. Teaching and practice

Every module includes hands-on work, and staff leave class with something they can use.

4. Follow-up

Collect common questions and arrange an advanced course if needed.

Deliverables

Course plan
Handouts and prompt templates
In-class results (such as a department assistant)
Post-course survey and feedback summary

Engagement cycle

Each cycle runs three months, six months or a year, depending on scope. At the end of each cycle we sit down with you and compare the results against the goals set at the start, then decide what the next cycle should cover, or whether to stop there.

Pricing

Each engagement is estimated on its own: how many systems and how much data are involved, the people and time needed, and how long the cycle runs. It comes in two steps. First we measure the baseline — before any quote, and priced separately — recording where things stand now. Then we estimate the rollout from the scope we measured, rather than quoting a price and fitting the scope to it. Write to us about where you are and we’ll explain what measuring the baseline involves.

Common questions

Can staff who have never used AI take the course?

Yes. We group staff by level and start from the basics before moving to structured prompting.

Which AI tools does the course use?

Mainly Google’s Gemini and NotebookLM. The methods apply to other AI tools too, and we adjust to the tools your company uses before the course.

What do we take away from the course?

Every module includes hands-on work. By the end, staff leave with prompt templates and the department assistant they built in class.

Tell us where you are

Email us about where you are
We reply with what could work
Baseline first, then scope and estimate