Discover
2 weeks · fixed scopeWe map the system, the constraints and the risk, then hand you an architecture document and a delivery plan. You keep both even if you stop there — and some clients do.
How we work
How we run a project, which engagement model fits which kind of work, exactly where we use AI and where we deliberately do not, and the gates a change has to pass before it reaches your users.
Softrear runs a three-phase process — Discover, Build, Run — on one of four engagement models, with CI-enforced quality gates and a published policy on AI use in delivery.
We map the system, the constraints and the risk, then hand you an architecture document and a delivery plan. You keep both even if you stop there — and some clients do.
Demoable software every two weeks, your team in the repository from day one, and our test and review standards applied to every pull request — including yours.
We stay on the pager. Named SLOs, a real on-call rotation, a quarterly cloud cost review, and a handover plan for the day you want to run it yourself.
Process
Three phases, each with a defined output, a defined shape and a defined exit. You are never in a position where stopping means having nothing.
Engagement models
Four models. The right one depends on how much of the scope is genuinely known, and being honest about that at the start saves both of us a difficult conversation in month three.
AI in delivery
AI-assisted development is real and it changes some of our costs. It also introduces failure modes that are easy to hide. Here is exactly what we do, so you can hold us to it.
Quality & testing
A definition of done that lives in someone’s head is a definition of done that moves. Ours is in the pipeline, and the pipeline is the arbiter.
A senior engineer reads every one of these. If we’re not the right fit we’ll say so, and point you at someone who is.
No spam, no drip sequence. A senior engineer replies within one business day.