How engagements are shaped, how delivery runs, what quality and security mean in practice, how you can contract us — and what we will not take on.
A brief is not a scope. Every engagement starts with a short discovery that produces a problem statement, a measurable definition of done, and an estimate range with the assumptions written down — so you can see which assumptions move the number.
Each increment ends with something deployed to a real environment that you can use. A sprint that produces only a status update has not produced anything, and we treat that as a delivery failure rather than a reporting one.
On platform work we meet requirements with native capability wherever it stretches. Every custom decision goes into a log with the reasoning, because someone will be looking at it during an upgrade three years from now.
Architecture decision records are part of delivery, not documentation debt. The reasoning behind a choice is worth more later than the choice itself.
Code lives in your repositories, infrastructure is defined as code, and runbooks are yours from the day they are written. Exit assistance is in the contract from the start. A supplier who makes leaving hard is managing their retention with your risk.
If a process should not be automated, if a brief is unfillable at the rate offered, or if a platform is the wrong answer, we say so. It costs us work occasionally and saves both sides considerably more.
Durations are indicative. The structure is not — every delivery engagement runs through these five stages, scaled to size.
Short, paid, and outcome-owned by you — the artefacts are yours whether or not we do the build.
Deliverables
Environments, pipeline, auth and observability stood up before feature work, because these are expensive to retrofit.
Deliverables
Plan, build, demo, adjust. Scope changes are priced at the sprint boundary rather than absorbed silently.
Deliverables
Performance, security, accessibility and resilience validated against production-like data and load.
Deliverables
Elevated support through first live running, then a deliberate handover to your team or into our managed services.
Deliverables
The second list costs us work sometimes. We would rather lose a deal than win one on a claim we cannot stand behind.
These are not optional extras quoted separately. They are what we mean when we say a thing is finished.
Unit, integration and end-to-end automation running on every commit, with an agreed coverage floor on business-critical paths written into the definition of done.
SAST, dependency scanning and secret detection wired into CI from the foundations phase, not added before a launch review.
Automated checks in the pipeline plus a manual keyboard and screen-reader pass before release. Built into the component library rather than retrofitted.
Core Web Vitals measured on real user traffic with budgets enforced in CI, so regressions are caught at the pull request rather than in production.
Every change reviewed by someone who will have to maintain it. Approval without reading is treated as a process failure.
Structured logging, metrics and tracing instrumented during foundations, so the first production incident is not also the first time anyone looks for a dashboard.
Particularly relevant if you are regulated, or if offshore access is a question your security team will ask about.
Our engineers work inside your identity provider, your approval workflow and your change control — not alongside them.
We default to masked or synthetic data for testing. Genuine production data in lower environments requires a documented approval and a defined retention window.
Where data residency obligations apply, access can be restricted by geography or the engagement run entirely onshore. Settled at the start, not at the audit.
Ticketing, change records and audit trails are structural to the workflow, so control evidence can be produced on demand rather than reconstructed.
The commercial shape determines where risk sits. Getting this wrong is the most common cause of a soured engagement, so it is worth being explicit.
| Model | Commercial shape | Risk sits with | Best fit |
|---|---|---|---|
| Discovery | Fixed fee | We carry scoping risk | Deciding what to do |
| Dedicated pod | Monthly per pod | You own priorities | Continuous roadmap delivery |
| Fixed-scope project | Milestone payments | We carry estimate risk | Stable, well-defined scope |
| Time & materials | Rate, capped monthly | You carry scope risk | Exploratory or shifting work |
| Staff augmentation | Rate per person | You own delivery | Adding specific skills |
| Managed services | Monthly service fee | We carry SLA risk | Ongoing run and support |
Not sure which fits? We wrote about how to choose
From a three-week discovery through to a standing pod or a multi-application managed service. We are a small firm and will tell you plainly when something is beyond our current capacity rather than take it and staff it thinly.
Fixed fee, typically two to six weeks depending on scope. The artefacts are yours whether or not you continue with us — which is the point. A free discovery is a sales process wearing a different name.
Yes, after discovery. Before discovery a fixed price is either padded to cover unknowns or optimistic enough to become an argument later. Fixed price works well for well-defined scope, and badly when you expect to learn things during the build.
Every engagement agrees fixed hours when the India delivery team and your team are both working. Standups, demos and escalations happen inside it; focused work happens outside it. It is agreed per engagement rather than imposed as a global policy.
You do, from the first commit. Code lives in your repositories, and IP assignment is explicit in the contract. There is no SegaKIV framework you would be licensing.
Notice periods are short by design and exit assistance is written into the contract. Documentation, runbooks and infrastructure code are yours throughout, so leaving is a scheduling problem rather than a rebuild.
A short conversation is usually enough to tell whether we're the right fit. If we're not, we'll say so and point you somewhere better.
Or email us directly at reachus@segakiv.com