Process
Four stages, each ending in documents you own. You always know what has been decided, what is being built and what happens next.
The method
The same four stages run on every engagement, scaled to its size. A two-week integration and a nine-month platform build follow the same method.
1–3 weeks
We learn the business goal, the constraints and the systems involved: workshops with your stakeholders, a review of existing code and infrastructure, and an honest read of the risks.
You receive
Your involvement
A product owner and a technical contact, about four hours a week.
2–4 weeks
We compare options and write the decisions down before anything is built. The architecture dossier records context, constraints, options and trade-offs, and we review it with your team.
You receive
Your involvement
Architecture reviews with your technical leads.
Two-week iterations
We build in two-week iterations. Each ends in a demo, a written progress report and test evidence, and we re-forecast the estimate at every phase boundary.
You receive
Your involvement
Demos, backlog decisions and a weekly sponsor check-in.
Warranty, then optional SLA
We hand over code, documents and knowledge so your team can run the system without us — and, if you want, we stay on under an SLA.
You receive
Your involvement
Your operations team joins the runbook drills before go-live.
Documents you own
These documents live in your repositories and are written for the team that will run the system after us.
Context, constraints, options and decisions — written before code and kept current.
One record per architecture decision, with the options considered and the trade-off accepted.
What will be tested and how, then the results for every release.
Deployments, rotations, backups and incident playbooks, rehearsed before go-live.
Monthly SLA performance, incidents, patches and improvements.
Engagement models
The method stays the same; the commercial model changes with how certain your scope is.
For work you can define up front.
Fits when
How it’s priced
Team
Cadence
For evolving scope and steady capacity.
Fits when
How it’s priced
Team
Cadence
For live systems that need an SLA.
Fits when
How it’s priced
Team
Cadence
You own — In every model: all code, documentation and intellectual property, from day one — in repositories you control.
Governance & support
Reporting is written down and predictable. After handover, support is a choice between three tiers, never a dependency.
Every iteration
Working software, test evidence and a plain summary of progress, risks and decisions.
Every week
Thirty minutes with your sponsor on priorities, blockers and anything that needs a decision.
Every phase
Estimate ranges updated against what we have learned, before the next phase is committed.
Always
Decisions, risks and open questions in a log you can read at any time.
Managed support tiers
SLA tier
For internal systems
SLA tier
For customer-facing systems
SLA tier
For systems that can’t stop
Tell us about the work. We’ll come back with a suggested first stage, the documents it produces and an estimate range.