Every engagement starts with a paid audit.
Projects fail in the first four weeks, before anyone writes production code, because the real shape of the work surfaces too late.
The four phases
PHASE 01
System audit
2–3 weeks · fixed fee
We map what you actually run, not what the documentation says you run. Interfaces, data flows, undocumented dependencies and the exceptions that have accumulated around them.
- Interface and data-flow inventory
- Code and infrastructure review
- Stakeholder interviews across business and IT
- Risk register with severity and likelihood
You receive
- Architecture map as-found
- Prioritised risk register
- Costed delivery plan
- Go / no-go recommendation
PHASE 02
Foundation
3–5 weeks
Before anything migrates, the ground has to hold. Environments, pipelines and observability go in first, then the first working slice runs beside the existing system rather than replacing it.
- Environment and pipeline setup
- Test harness on the affected surface
- First vertical slice in production
- Rollback procedure proven, not assumed
You receive
- Running CI/CD pipeline
- Staging environment
- First slice live
- Rollback runbook
PHASE 03
Delivery
Ongoing · two-week cycles
Increments against the plan, with a demo at the end of every cycle. Scope changes are re-costed in the open rather than absorbed silently and surfaced at the end.
- Two-week increments with demo
- Written cycle report and burn-up
- Open change control on scope
- Continuous integration into your systems
You receive
- Working software each cycle
- Cycle reports
- Updated plan and cost view
- Access to the repository throughout
PHASE 04
Handover
2 weeks + SLA
The engagement ends with your team able to run the system without us. Documentation is written for the engineer who arrives next year, and onboarding is paired rather than presented.
- Paired onboarding with your engineers
- Runbook and incident procedure walkthrough
- Architecture decision records handed over
- Optional support retainer agreed
You receive
- Technical documentation
- Operational runbooks
- Decision record archive
- Named SLA if retained
Engagement models
Three ways to contract the work.
Selected at the end of the audit, when scope is known well enough for the choice to be honest.
Fixed-fee audit
Always first
A defined scope, a defined price and a defined output. The only way we start, and the only commitment required to begin.
Capped scope
Known deliverable
For work the audit has defined clearly: a compliance integration, a reporting model, a platform migration with a fixed end state.
Retained team
Ongoing programme
A standing engineering capacity with named people and named SLA tiers, for organisations with a continuous roadmap rather than a project.
Assurance
What procurement asks before legal does.
On the table from the first call, not extracted over six weeks of vendor review.
- Contracting
- Fixed-fee audit, then time and materials or capped scope. Your paper or ours.
- IP ownership
- You own the code and the repositories from the first commit, without escrow clauses.
- Data handling
- GDPR and KVKK aligned. Data residency in your region, DPA signed before access.
- Security
- SSO, least-privilege access, dependency scanning and secrets management as standard.
- Continuity
- Two engineers minimum on every codebase. No single point of knowledge failure.
- Support
- Named SLA tiers with response targets measured in hours.
Support tiers
Standard
For systems with a stable release cadence and internal first-line support.
- Response
- Next business day
- Coverage
- Business hours
- Channel
Priority
For revenue-facing integrations where a failure blocks transactions.
- Response
- 4 business hours
- Coverage
- Extended hours
- Channel
- Email + shared channel
Critical
For compliance filings and platforms with a regulatory deadline attached.
- Response
- 1 hour
- Coverage
- Extended + on-call
- Channel
- Named engineer
Start with the audit.
Fixed fee, two to three weeks, and the output is yours whatever you decide next.