How we work
Working software every week, decisions in writing.
The process is deliberately plain. It exists so you always know what is being built, why, and what it will cost to change your mind.
Process
Assess, architect, build, deliver.
- 01
Assess
We read the code, talk to the people who use and run the system, and write down what the outcome has to be. Constraints, integrations and unknowns are listed, not assumed.
You get
- Written scope and estimate
- Risk list with proposed mitigations
- Go / no-go decision with no obligation
- 02
Architect
Architecture, data model and interfaces are agreed before implementation. Where the risk is technical rather than conceptual, we build a thin prototype instead of a document.
You get
- Architecture and data model
- Prototype for the riskiest part
- Milestone plan
- 03
Build
Work ships into your repository in small, reviewed pull requests. Each week ends with something you can click on or query, plus a short written update.
You get
- Working software every week
- Tests, CI/CD and documentation
- Weekly written status
- 04
Deliver
After launch we monitor, fix what reality reveals and hand over. Continue with a retainer, or take it fully in-house; the codebase is built for either.
You get
- Monitoring and runbooks
- Prioritised backlog of next steps
- Handover session with your team
Engagement models
Three ways to engage.
Pick the shape that matches the problem. Many clients start with a scoped project and continue on a retainer.
Scoped project
Fixed scope, fixed outcome
A defined deliverable with an estimate and milestones: a migration, an internal tool, a data pipeline, an AI workflow in production.
Best for: a clear problem with a deadline
Engineering retainer
Reserved capacity each month
Ongoing development and operations for a product or platform, planned in a shared backlog with weekly check-ins.
Best for: a product that keeps evolving
Technical advisory
Senior judgement on call
Architecture reviews, audits, hiring support and AI adoption planning for founders and teams who build in-house.
Best for: decisions, not delivery
What to expect
Rules we hold ourselves to on every engagement.
- Your repository, your accounts
- Code, infrastructure and credentials live with you from day one. There is nothing to hand over at the end because nothing was ever ours.
- Weekly, visible progress
- A short written update and a working increment every week. If something slips, you hear it from us first, with options.
- Tests and docs are part of done
- Nothing ships without automated tests, a deployment pipeline and enough documentation for the next engineer, whoever that is.
- Boring technology by default
- Proven tools your team can hire for. New technology, including AI, earns its place by solving a problem you actually have.
- Senior people, direct line
- You talk to the engineers doing the work. No account layer, no translation loss.