Process
Two weeks to your prototype. Six weeks to your system.
Fixed scope. Fixed price. Honest timelines for a two-person studio. One active build at a time.
Phase 01
Founder MVP · 2 weeks
We cut your idea to its smallest useful version, then ship it with auth and a real database.
- Day 01
Scope
60-minute working session. We agree on the smallest version of your idea that proves value, then create a one-page spec.
- Goal, users, and the one scenario the prototype must handle
- Stack chosen on the call (no fashion projects)
- What is in scope, what is explicitly out
- Days 02–07
Build
Repo set up, infra wired, the core loop running end-to-end with placeholder data.
- Project repo, env templates, deploy pipeline
- Core user journey, page flow, or automation working end-to-end
- Twice-weekly Loom showing what is alive
- Days 08–12
Iterate
You try it. We sharpen. Edge cases caught and named, with a tiny eval suite for the critical path.
- Real data plugged in
- Edge cases caught and named
- Tiny eval suite for the critical path
- Days 13–14
Ship
Hosted, documented, demoable. Either we stop here or roll into the System phase.
- Hosted demo link
- Source, prompts, and configs handed over
- 30-min review call: keep, scale, or kill
Phase 02
System · 6 weeks
We harden your prototype into a production system you can run for years.
- Weeks 1–2
Foundation
Auth, roles, data model, and the first integration. Everything typed end-to-end.
- Auth and roles wired
- Data model finalised
- First integration live (CRM / payments / telephony)
- Weeks 3–4
Depth
Real flows for the people who will use this every day. Two more integrations as needed.
- Workflow UIs
- Background jobs and queues
- Up to two more integrations
- Week 5
Hardening
Quality gates, tracing, monitoring, and the boring-but-important things that keep production systems honest, including evals when AI is in the loop.
- Tests or evals for critical paths
- Tracing and structured logs
- Performance, cost, and latency monitoring where relevant
- Week 6
Launch
Deployment, documentation, handover, and two weeks of bug-fix support.
- Production deployment and runbook
- Documentation and handover session
- Two weeks of post-launch bug fixes
Guarantees
No drama, by design.
Scope locked on Day 1
You get a one-page spec and fixed price before we write code. Timelines are estimates for our two-person studio.
You own everything
Code, prompts, evals, and weights go to you on delivery. No lock-in, no per-seat fees.
No surprise scope
Scope is locked on Day 1. Any addition is a separate, written change order. No silent creep.
FAQ
Questions about your timeline
What happens on Day 1?
You join a 60-minute scoping call. We lock scope, pick the stack, and send a one-page spec with fixed price before any code is written.
Do I need a detailed spec before we start?
No. You bring the problem and examples. We shape the spec together on Day 1.
What if the MVP takes longer than 2 weeks?
We set honest timelines for a two-person studio. If something slips, we tell you early and agree on next steps in writing. We do not publish refund guarantees.
Can I pause between MVP and System?
Yes. The MVP is yours either way. You decide when to continue.
How do updates work during the build?
You get a Loom twice a week showing what runs. You test mid-build and send feedback in writing or on a quick call.
How many projects do you run at once?
One active client build at a time. Both of us stay on your project until handover. Expect 1–2 weeks before kickoff if we are booked.
Bring your problem. We bring the playbook.
Tell us what you want to build on the intake form. We reply within one business day with a scoping slot.