Skip to content
New guide: how to scope an AI project before you build

Intelligence

Intelligent Software Systems

Systems that combine your data, rules and models into software your team can actually run and change.

Typical timeline
6-12 weeks
Group
Intelligence
First step
Fixed-price discovery

Scope, price and acceptance criteria are agreed in writing before the build starts.

Why us

Why choose us for this work?

Data model first

We model the entities and rules before choosing how intelligence is applied.

Fits the process

The system is designed around how the work happens, not around a demo flow.

Owned by you

Code, data and decisions live in your repositories and accounts.

Testable

Behaviour is captured in tests and evaluation sets that run on every change.

Maintainable

Mainstream languages and written decisions keep the next change cheap.

What you get

What you get from Intelligent Software Systems

A system, not a prototype.

  1. 01

    A working system

    One end-to-end capability running against real data.

  2. 02

    An explicit data model

    Entities, rules and ownership written down and enforced.

  3. 03

    Evaluation and tests

    A repeatable measure of quality that runs in CI.

  4. 04

    A usable interface

    Screens built for the people who will run the process.

  5. 05

    Documentation

    Architecture decisions, runbooks and handover notes.

The problem

What usually goes wrong

A model on its own is not a system. Useful intelligence needs a data model, permissions, evaluation, a user interface and a way to recover when something fails. Those are software engineering problems.

Who this is for. Teams that need intelligence embedded in an existing product or operation, not a separate demo.

Signs you need this

If any of this sounds familiar

  • Important knowledge is spread across spreadsheets and inboxes.
  • People make the same judgement many times a day.
  • A prototype exists but has no path into production.
  • You need the system to fit an existing process.

How the work runs

How we deliver it

  1. 01

    Understand

    We map the data, the decisions and the people involved.

  2. 02

    Agree the measures

    We define what better looks like before building.

  3. 03

    Build in weekly steps

    Each stage produces something testable and demoable.

  4. 04

    Test against reality

    Real examples, missing data and edge cases are used throughout.

  5. 05

    Hand over

    You keep the code, the data and the decision log.

Deliverables

What you receive

  • Domain and data model
  • Application with an integrated intelligence layer
  • Evaluation harness and automated tests
  • User interface for the operating team
  • Runbooks and architecture decisions

Stack

Tools we expect to use

TypeScriptPythonPostgreSQLReactCloudflare

We choose the stack to fit the problem. If your team already runs something else, we plan around it.

Questions

Common questions

Will it work with our existing tools?

Usually yes. We plan around your systems of record rather than replacing them, and document the boundaries clearly.

Can our team maintain it?

That is the goal. We use mainstream languages, write decisions down and pair during handover.

What if the model changes?

The model sits behind an interface. Your data, rules and evaluation stay yours, so swapping it is a contained change.

Next step

Talk through your project.

A short conversation. You leave with a clear view of what it would take, and whether this service is the right first step.

Prefer email? Write what you are working on and we will reply.

Send a note

By sending this, you agree that we may reply to your enquiry. See our privacy policy.