KeevaathTech

KeevaathTech/How we work

How we work

Four stages, the same order every time

Every engagement moves through the same four stages, on client builds and on our own platforms. Scope changes and the order does not.

Four stages, the same order every time

Every engagement moves through the same four, on client builds and on our own platforms. Scope changes. The order does not.

Four stage sequenceDefine, design, build and run shown as four rising platforms with progress flowing between them and operation returning as a loop.DEFINE01DESIGN02BUILD03RUN04
01 · Define · 1 to 2 weeks

What the product needs to do, who it is for, and what success looks like, written down before anyone writes code.

02 · Design · 2 to 4 weeks

Interface and architecture together. Data model, integrations and hosting decided alongside the screens, because deciding the screens alone produces screens that cannot be built.

03 · Build · 6 to 16 weeks

Working software in short cycles. You open it running in a real environment every cycle, not a status report describing one.

04 · Run · ongoing

Launch, monitor, fix and extend, with handover documentation if you are taking it in house.

Who actually does the work

A named owner at KeevaathTech runs every engagement and stays your point of contact throughout. That person owns the architecture, the quality bar and the schedule.

The team scales to the engagement, with specialists matched to the problem instead of assigned from whoever happens to be free, and every piece of work passes a review here before you see it.

Part of our delivery capacity works Indian Standard Time, which gives roughly four to five hours of daily overlap with United States Eastern. We publish that because it is the first thing worth knowing and the last thing most firms will tell you.

Inside the platforms

Our platforms do not stop at answering a question. They reason across a request, call the tools they need, check their own output, and hand a person something they can act on. Every step is recorded, so when a result is wrong we can see where it went wrong.

Agentic run pipeline Six stages standing on a rail, with data travelling between them and a self check stage that loops back to correct earlier output. Hover or focus a stage to highlight the matching line in the run log. INGEST 01 EXTRACT 02 CHECK LAW 03 FLAG · RANK 04 SELF CHECK 05 RETURN 06 DROP UNSUPPORTED FLAGS

The interesting engineering is not the model. It is step five, and the logging that makes step five auditable. A system that cannot show its work is not one we would put in front of somebody deciding whether to take a job in another country.

The same architecture is what clients get. How we build it.

How we quote

Fixed scope work

Assessments, audits, evaluations and the discovery sprint are quoted at a fixed price against a fixed end date. You know both before anything begins.

Build and strategy

CORE and FRAME are quoted after discovery, because an estimate made before the problem is understood is only a guess with a decimal point on it.

Ongoing work

RUN maintenance is quoted on response time, not on hours logged, and always after the audit that tells us both what we are agreeing to.

Turnaround

A completed questionnaire usually comes back with a quote within two working days, from the person who would run the engagement.

We do not publish rate cards, because the same audit across two templates and across forty is not the same job, and a number that ignores that helps nobody.

What we will tell you early

If we think something should not be built, or should be built by somebody else, we say so while it is still cheap to hear. Roughly one discovery sprint in five ends with a recommendation not to build.

If your team inherits the codebase it should be legible without us. Documentation and a clean handover are part of the build and never an upsell. We would rather you could leave than rely on the fact that you cannot.

Questions we are asked

How long before we see something working?

On a build, the first cycle of week three or four. You open it running in a real environment, not a demonstration recorded for you.

Can our team work alongside yours?

Yes, and it is often right when the plan is to take the product in house. We agree who owns which parts up front so review and merge do not become a negotiation.

What happens if we want to stop?

You take the code, the documentation and the environment. It is written into the engagement, because a studio that makes leaving expensive is relying on the wrong thing.

Who is our point of contact?

One named person, who quotes the work, runs it and is still there at the end. Not an account manager relaying messages.

Do you work fixed price or time and materials?

Fixed price wherever the scope can genuinely be fixed, which is every assessment and the discovery sprint. Builds are quoted after discovery, because a fixed price on an undefined scope protects nobody.

Tell us what you are building.

Send the shape of the problem: scope, timeline and constraints. You do not need a specification or a budget worked out.

Related