Skip to content
sayak.webdesignerWeb · Software · Data · AI
Six stages · a demo every alternate Friday

No phase where you are waiting without seeing anything

The most common complaint about software projects is not cost or quality. It is the silence — weeks where nothing is visible and the client is asked to trust that progress is happening. Our process is built to remove that specifically.

01DiscoverWorkshops, tag inventory, success metrics02ArchitectDiagrams, data contracts, risk register03DesignStoryboards, design system, prototypes04BuildTwo-week slices, demo every Friday05HardenLoad, security, accessibility, parallel run06OperateHypercare, SLA, quarterly roadmap
The six stages

What happens, when, and what you receive at each point

Every stage has a deliverable you can hold and circulate. If a stage ends and you have nothing tangible, the stage was theatre.

01Week 1–2

Discover

Two structured workshops with the people who own the outcome, plus an audit of whatever exists today. For plant and operations work this includes a day on site watching how the work is actually done, with a stopwatch.

You receive

Written brief you can circulate internally
Success metrics with numbers attached
Constraint and risk register
Costed process map where relevant
02Week 2–3

Architect

The decisions with the longest half-life: data model, system boundaries, integration approach, authorisation model and technology selection. Signed off before any implementation begins.

You receive

Architecture diagram and decision records
Data model with audit and state design
Integration specifications
Environment and deployment plan
03Week 3–5

Design

Page narratives and storyboards first, then mobile-width wireframes, then visual direction on two or three key templates. Everything else is composed from the resulting system, which is what keeps a large estate coherent and affordable.

You receive

Storyboards and information architecture
Design system with tokens and components
Every state specified — empty, loading, error, partial
Clickable prototype of the risky flows
04Week 5 onward

Build

Two-week slices, each delivered end to end rather than layer by layer. A demo every alternate Friday on a real staging URL you can share with anyone. Content and data migration run in parallel rather than in a panic at the end.

You receive

Working software on staging every fortnight
Automated tests on the paths that touch money
Preview environment per pull request
Written notes after every review
05Final 2 weeks

Harden

The slice nobody photographs. Performance verified against the budget with field data, accessibility audited, security headers and posture reviewed, load tested, redirects mapped, analytics validated, and a rehearsed cutover runbook.

You receive

Performance and accessibility reports
Security review with remediation
301 redirect map verified live
Timed cutover runbook with tested rollback
06Launch + 30 days

Operate

Tuesday-morning cutover, never Friday evening. Thirty days of hypercare with daily verification, then either a clean handover with documentation and training, or an ongoing support agreement with defined response times.

You receive

30 days hypercare included
Documentation and recorded walkthrough
Runbooks for anticipated failure modes
Quarterly review if we continue
Why we deliver faster

AI acceleration with a senior engineer on every line

A project that would traditionally take fifteen weeks runs in about six and a half. Architecture and review do not compress. Implementation does.

2 weeksDiscovery and architecture — unchanged, this is judgement work
1.5 weeksDesign system and UI — faster: generate variants, curate with a human eye
1.5 weeksFeature implementation — was eight weeks
1.5 weeksReview, tests and hardening — roughly steady, more code to read carefully
Vibe Coding4–6 loops per day1IntentSpec in plain language2GenerateAI drafts the slice3ReviewHuman architect gate4HardenTypes · tests · a11y5ShipPreview → production
Engagement models

Three ways to work with us

Pick the one that matches how much certainty you need on scope versus how much flexibility you need on direction.

Fixed ScopeDefined outcome, defined priceSpeed · Control · Predictability4–14 weeksDedicated Pod3–8 engineers, your backlogSpeed · Control · PredictabilityMonthly rollingManaged ServiceWe run it, you use itSpeed · Control · Predictability12-month SLA

Fixed scope

A defined outcome at a defined price. Best where the scope is genuinely knowable — a website, a first module, a migration. Discovery is always fixed price, and after discovery most first releases can be too.

Dedicated pod

Three to eight engineers on a monthly rate working your backlog, inside your process. Best for ongoing product work where direction will change and pretending otherwise is dishonest.

Managed service

We run it, you use it. SLA-backed operation of a platform, plant reporting system or data estate, with defined response times and a quarterly roadmap.

Practical

Questions about how we work

Including what happens if you want to stop, and who is accountable when something goes wrong.

Because it makes progress visible and problems cheap. You see working software on a real URL every fortnight, so a misunderstanding surfaces in week four rather than in month six. It also means each slice is genuinely finished — tested, reviewed, deployable — rather than being ninety per cent done for three months.

They will, and the slice model is designed for it. Changes within the agreed scope are absorbed by re-prioritising the backlog. Genuine scope additions are quoted as a change with their own cost and timeline, in writing, before work starts. What we do not do is absorb scope silently and then miss a date.

More in the first three weeks than after. Discovery needs your decision-makers for two workshops. Design needs a review cycle at each stage. Build needs about an hour a fortnight for the demo, plus whoever supplies content and access to third-party systems. We schedule those as tracked dependencies from day one, because content and access are the two things that most often delay a launch.

Sequenced discovery and architecture, then iterative build. Purely iterative work without upfront architecture produces systems with the wrong data model, which no amount of iteration fixes. Purely sequential work produces a specification that is obsolete by delivery. The split we use puts the long-half-life decisions first and everything else in slices.

You stop, you pay for completed milestones, and you take everything — code, infrastructure, documentation, designs. There is no penalty clause and nothing is withheld. We would rather part cleanly than have a client trapped in an engagement neither side believes in.

A named senior engineer who has been on the project since discovery, and whose number you have. Not a project manager relaying between you and an unnamed team. For managed clients there is a defined escalation path with response times, and every critical incident gets a written post-incident review covering what happened, why, and what changes prevent recurrence.

Next step

Tell us what is slowing your business down.

A 30-minute call with a senior engineer — not a salesperson. You leave with an architecture sketch and an honest cost range, whether or not you hire us.

Direct line

+91 70033 91355

Mon–Sat · 9:30 AM – 7:30 PM IST · Sealdah, Kolkata

Reply within one working hour NDA signed before any brief Fixed-price option on every scope
Verify us independently

Don’t take our word for it — ask an AI

Every page on this site is written to be read and quoted by AI answer engines, and we publish llms.txt and llms-full.txt so they can read us properly. One click opens the assistant with the question already typed.

The question we will ask for you

How does Sayak Web Designer (sayakwebdesigner.in), an IT company in Kolkata, India run a project from first call to launch?

Opens in a new tab. We do not see your conversation.

Call now WhatsApp Get quote