AI-assisted prototyping for web platforms
A working prototype of your platform, built with AI assistance and checked by a developer. It looks and works the way the finished platform should, so you can test the workflows and find problems before the engineering work begins.
A working prototype before the engineering starts
Before a platform is engineered, we build a prototype of it with AI coding assistants. It is not a set of pictures: the screens, forms, workflows and user roles work the way the finished platform is meant to.
You can click through it, try it with sample data and show it to the people who will use it. Problems that would be expensive to fix later show up while they are still quick to change.
Why prototype first
Most problems in a platform are cheaper to find before it is built than after.
Work out the workflows
Follow each task from start to finish, such as an enquiry becoming a quote and then a job, and see where steps are missing or in the wrong order.
Test with sample data
The prototype is filled with sample data that looks like yours, so you can test it properly without using real customer records.
Find bugs early
Edge cases, missing fields and confusing rules come up while the prototype is being tested, not after launch.
Catch design flaws
See whether screens make sense to the people who will use them every day, and change them before they are engineered.
Agree what gets built
Everyone looks at the same working version, so the engineering work starts from something agreed rather than a long document.
Show it to others
Use it to explain the idea to your team, users or partners, and collect their feedback early.
What a prototype can include
We prototype as much of the platform as you need to test, from one workflow to the whole system.
Screens and navigation
Every main screen, laid out and linked the way the finished platform will be.
Forms and records
Forms that save, records you can search, edit and change status.
Workflows
The steps and statuses a task moves through, with what happens at each one.
User roles
Different people see and do different things, for example staff, managers and customers.
Sample data
Enough realistic sample data to see how lists, searches and dashboards behave.
Dashboards and reports
The figures and lists people need to see, built from the sample data.
How AI helps, and where a developer decides
AI coding assistants let us build screens, forms and sample data much faster than by hand, so more of the platform can be prototyped and changed between rounds of testing.
A developer plans the structure, checks what the AI produces and decides what goes in. It is the same way we work on AI-assisted engineering.
How we prototype our own platform
We prototyped our own platform this way. The Bones 9 staff console and customer portal were specified and prototyped before the engineering work, on top of the Bones 9 core platform.
Bones 9 is our own work. It is still in development and not generally available.
How it works
Every prototype is scoped and priced before work starts.
-
1
A short call
We talk through what the platform is for, who will use it and which parts you most need to test.
-
2
Scope and quote
We agree what the prototype covers and the price, before anything is built.
-
3
Build and test
We build the prototype, you try it with sample data, and we change it between rounds of testing.
-
4
Ready for engineering
You end up with an agreed working version and a list of the bugs, gaps and design changes found. The engineering work starts from there.
-
5
The real build
The finished software is written new, not copied from the prototype. Our engineers work from the prototype as the reference for the rules, workflows and styles, which makes the build quicker and clearer.
Prices
Prototypes are quoted after a short call, because the price depends on how much of the platform the prototype covers. You can add one to an indicative quote on the pricing page.
FAQ
Common questions
Short answers to what clients ask us most. Anything else, just ask.
Ask a question
What is an AI-assisted prototype?
A working version of your platform, built quickly with AI coding assistants and checked by a developer. It looks and works the way the finished platform is meant to, so you can click through it and test it before the engineering work begins.
How is it different from a design mock-up?
A mock-up shows what screens look like. A prototype also works: forms save, records change status, different users see different things, and the workflows run from start to finish.
Is the prototype the finished platform?
No. It is built to test ideas, workflows and design. Once it is agreed, the engineering work makes the platform ready for real users and real data.
Is the prototype code used in the finished software?
No. The finished software is written new. The prototype is the reference our engineers build from: it shows the rules, workflows and styles that were agreed, which makes the real build quicker and clearer.
Do we need to give you real data?
No. We fill the prototype with sample data that looks like yours, so you can test it properly without using real customer records.
Have you built prototypes like this before?
Yes, for our own platform. The Bones 9 staff console and customer portal were specified and prototyped before engineering. Both are still in development and not generally available.
How is it priced?
A prototype is quoted after a short call, because the price depends on how much of the platform it covers. The price is agreed before work starts.
Insights
Related insights
Articles related to this service.