How we price
A fixed price against a written scope, agreed before anything starts. No hourly surprises, no lock-in.
The short version
You get a number before you commit to anything
Tell us the problem and we come back within one business day with an honest read on whether we can help. If it fits, we scope it properly and send a written price and a date. That stage is free and carries no obligation.
There is no price list on this page on purpose. Two projects described in the same sentence routinely differ by an order of magnitude, and a published range that does not apply to you is worse than no range at all. What we can be specific about is the model, and what moves the number.
Always included
- Written scope, fixed price and a date before work starts
- Weekly demos against the agreed scope
- Deployed into your cloud account, not ours
- Full handover of code, credentials and documentation
- No proprietary licence between you and your system
- Response within one business day
Engagement models
Three ways we work
Most first projects are fixed-price. The other two exist because some work does not have an end state.
Fixed-price project
A defined build with a clear end state
- Written scope with a fixed price and a date, agreed before work starts
- Weekly demos against that scope
- Change requests priced separately rather than absorbed silently
- Most common for a first engagement
Retained capacity
Ongoing work where the backlog outlives any one project
- An agreed block of senior engineering time each month
- Priorities set by you, re-set as often as you like
- Suits teams extending a system we already built
- Cancellable — you are not buying a lock-in
Support and maintenance
A system that is live and needs to stay that way
- Monitoring, patching, backups and certificate renewal
- Quoted separately from any build, never bundled into it
- Optional in every case — the handover stands on its own
- You can take it in-house whenever you want
What moves the number
Five things that decide the price
Ask any supplier about these. The answers tell you more than the quote does.
How many systems it touches
Integration count drives more of the price than feature count. The cost is set by the worst-documented system in the list, not the average one — a single tool with no API can outweigh everything else.
Whether it is a build or a rebuild
Taking over an existing codebase starts with reading it and telling you honestly what is worth keeping. That is cheaper than a rewrite when the foundation is sound, and more expensive when it is not.
How wrong it is allowed to be
Confidence scoring, review queues, deterministic guardrails and evaluation against your own test cases are real engineering. A prototype skips all of it; anything customer-facing cannot.
Where it has to run
Deploying into your own cloud with reproducible builds, monitoring and a tested restore is more work than pushing to a hosting platform — and it is what makes the handover real.
Who operates it afterwards
Documentation written for whoever inherits the system, and the handover session to go with it. Optional ongoing support is quoted separately, never bundled to inflate the build.
A longer treatment of this, for AI projects specifically, is in what actually drives the cost of an AI agent build.
Engagement
How a project runs
The same five steps whichever service you pick.
- 01
Discover
Tell us what you're trying to solve. Every submission is reviewed personally — no bots, no gatekeeping.
- 02
Design
A no-obligation scope with a clear plan and the next steps, back within one business day.
- 03
Build
Senior engineers who've shipped at scale build the risky part first, with weekly demos against the scope.
- 04
Deploy
Deployed to your cloud, documented and transferred — you own the code and the infrastructure.
- 05
Support
Ongoing monitoring and technical support, for as long as you want us to stay on.
Questions
About cost and terms
The commercial questions people ask before the technical ones.
Get free consultationServices