01
Custom software
Web applications, internal tools and integrations for companies that will still be using them in five years. We design the architecture, write the code, put it into production and stay on to maintain it.
- Web applications and customer portals
- Internal tools that replace spreadsheets and manual workflows
- Integrations between ERP, accounting, CRM and logistics systems
- APIs and back-ends for mobile and partner applications
- Taking over and stabilising existing codebases
- Long-term maintenance and support
02
Infrastructure & consulting
Servers, networks and hardware sized to the load you actually have. We audit what's there, plan what's needed, roll it out and keep it healthy, whether it runs on-premise, in the cloud or both.
- Infrastructure audits and capacity planning
- Server, storage and network design
- Choosing, installing and maintaining hardware
- Cloud, on-premise and hybrid setups
- Backups, monitoring and disaster-recovery planning
- Ongoing administration and support
03
AI in existing systems
AI built into the software and infrastructure you already run, so there's no new platform to adopt. Where data can't leave your premises, the models run on your own hardware.
- Document processing: invoices, contracts, customs forms
- Natural-language search over internal documents
- Support and ticket triage
- Self-hosted models, including sizing and setting up the hardware
- AI-assisted modernisation of legacy code
- Readiness audits: where AI pays off, and where it doesn't
AI in practice
Not a chatbot on your website
Self-hosted AI on your own servers
Open-weight models running on infrastructure you control. Useful for banks, healthcare and the public sector, where data can't go to a third-party cloud.
Ask your old system a question
Natural-language search over documents, records or an internal wiki you already have, without moving off it.
Legacy code, modernised faster
AI-assisted tooling that speeds up migration and refactoring of older codebases. Same engineering discipline, shorter timeline.
An honest answer on where it pays off
We look at your operations and systems and point out the two or three places where AI would actually make a difference.
How we work
Four steps, every project
Listen
We start with how your systems and people work today, before we propose anything.
Plan
A written proposal with scope, architecture and cost, agreed before any code is written.
Build
Short iterations you can see and test, so there are no surprises at the end.
Support
Maintenance, monitoring and support after launch, for as long as the system runs.
Delivery
How projects are run
- A named technical lead on every project
- Documented architecture
- Source code and documentation handed over to you
- Peer review of every change
- Staged releases
- Monitoring and backup planning
- A maintenance agreement after launch
Security and data location
Security and data location are part of the architecture from the start. Access, backups, logging and deployment boundaries are agreed with you before implementation begins.
Not sure where to start?
Tell us what you're dealing with. Working out which of the three you need is part of the job.