AI coding assistants for development teams.
Your developers learn to work with an assistant on your own codebase, with review habits that keep quality where it is.
Who it’s for
- Developers
- Tech leads and architects
- IT and data teams
Formats
| # | Format | What it is | For this audience |
|---|---|---|---|
| Discovery half-day | A first look, on your own examples. | Not offered for this audience | |
| Team day | One team, one day, its own tasks. | Applies | |
| Multi-session programme | Several sessions over a few weeks. | Applies | |
| Open inter-company session | Several companies in one room. | On request |
What participants leave with
- 01
Your codebase, ready for the assistant
Context files on your architecture, conventions and commands, kept with the code.
- 02
Review habits that hold
Every generated change read, tested and reviewed by a developer before merge.
- 03
Team rules for AI coding
Which code and data may be shared, and where the assistant must not go.
Built on your files
Before the session, we write the context files for your repository: architecture, conventions, commands. They stay with you.
How the context file works
Context files are plain text kept in the repository. They tell the assistant how the project is organised, how to build and test it, which conventions apply and which areas are off limits. Participants work with them on real tickets from your backlog. After the training they are versioned with the code and maintained by the team.
Use cases by sector
Real estate
- Lead imports from listing portals cleaned and deduplicated
- An internal tool for tracking buyer files, prototyped and reviewed
- Checks on listing data scripted before publication
Manufacturing
- Legacy scripts around plant systems explained and documented
- Tests written before changing production reporting code
- Data exports from plant systems automated and tested
Finance
- Reconciliation scripts reviewed and covered by tests
- Old reporting code documented before a migration
- Code reviews focused on calculations and edge cases
Trading
- Market-data scripts refactored with tests
- Small internal tools prototyped and reviewed before use
- Documentation written for scripts only one person understood
Consulting
- Data-cleaning notebooks turned into maintained code
- Client prototypes built and reviewed before delivery
- Reusable project templates with tests and context files
Services
- Document-processing scripts tested on sample files
- Internal dashboards built from existing exports
- Small integrations between existing tools, reviewed before use
After the session
Two days later, a written recap: what was built, what worked, and the use cases not yet covered. An optional monthly follow-up keeps adoption on track.
-
End of the session
Something that works
-
Two days later
Written recap
-
Each month, optional
Adoption follow-up
Contact
Plan a session for your developers.
Related
Keep exploring.
Start with a 20-minute conversation.
Tell us where you want AI to make a difference first. No commitment at this stage.
Request a first conversation