Process

How an engagement actually runs.

Five phases. Each one has a written deliverable. Each one has a stop point where you can decide not to continue. Software engineering is risky enough — the engagement structure shouldn't add to it.

01

Discovery

Week 1

We start with a paid discovery week. The deliverable is a written report — not a deck — that lays out what we found, what the actual problem is (frequently not what was stated), the realistic scope, the risks, and a recommendation. If the recommendation is "don't do this," we say so.

Artifacts

  • ▸ Discovery report
  • ▸ Risk register
  • ▸ Realistic scope and budget
02

Architecture

1–3 weeks

If the engagement proceeds, we design the system before we build it. Service boundaries, data shapes, integration contracts, security model, observability plan, deployment pipeline. We get sign-off on the design before code lands.

Artifacts

  • ▸ Architecture decision records (ADRs)
  • ▸ Sequence and component diagrams
  • ▸ Schema and API contracts
03

Build

Engagement-dependent

Two-week iterations against a versioned scope. Working code at the end of each one, demoed to whoever can authorize change. Trunk-based development, code review on everything, automated CI from day one.

Artifacts

  • ▸ Working software
  • ▸ Test suite
  • ▸ CI/CD pipeline
  • ▸ Bi-weekly demo
04

Ship

1–2 weeks

Production deploy is a planned, rehearsed event. Feature flags. Canary or staged rollout. Rollback path documented and tested before we cut over. Whoever is on call sees the dashboard.

Artifacts

  • ▸ Production runbook
  • ▸ Rollback plan
  • ▸ Day-one monitoring dashboard
05

Hand-off

1 week

Knowledge transfer is part of the deliverable. We write the runbooks. We update the diagrams to match what shipped. We sit in on the first on-call rotation if you want us there. The system is yours, fully, on the day we step back.

Artifacts

  • ▸ Runbooks
  • ▸ Diagram set
  • ▸ Decision log
  • ▸ On-call shadow
06

Support (optional)

Month-to-month

If you want us on retainer afterward — to handle production incidents, evolve the system, or back up your team — we do that. Otherwise, the engagement ends cleanly at hand-off.

Artifacts

  • ▸ SLA-backed retainer
  • ▸ Quarterly architecture review
  • ▸ Incident response

How we behave

Four commitments we make in every contract.

Fixed scope, fixed timeline, written down.

Every engagement starts with a written scope. Changes go through written change requests. Nothing is informal. This protects you and us.

You own the source code from day one.

All work is delivered into your repositories from the first commit. No vendor lock-in. No "give it back when we leave." If we walk tomorrow, the code is already yours.

We talk to whoever can make the call.

Engineering decisions need engineering decision-makers. We don't go through three layers of project management to ask whether a column should be nullable.

If we get something wrong, we fix it on our dime.

Defects in shipped code are our problem, not yours. We don't bill for fixing our own bugs.

Ready to start with a discovery week?

The first call is free. The discovery week is a fixed price. After that, you decide whether to continue — with a real plan in hand.

Get in touch