How we work

How an AI engagement works

Every engagement runs in short stages with something real at the end of each one, so you can stop or change direction at any point. We agree the price of each stage before it starts rather than publishing a rate card, because the cost depends on your workflows, your data and the systems in scope.

  1. 01Discover

    1 to 2 weeks

    Interviews with the people doing the work, a look at your data and systems, and a written plan: what to build first, what it needs, what it would cost.

    What you get
    An opportunity map, the data and guardrail requirements, two or three architecture options with their running costs, a prioritised build plan, and a prototype scope you could hand to any team.
    How it is priced
    A fixed price, agreed before we start, based on the number of workflows and systems in scope.

  2. 02Prototype

    2 to 6 weeks

    A working agent, MCP server or retrieval system on your real data, built to find out what breaks before it matters.

    What you get
    A working agent, MCP server or retrieval system on your real data. For an MCP server, a design sprint of about two weeks produces the tool design, the authentication design and a prototype against one host.
    How it is priced
    A fixed price, agreed after discovery.

  3. 03Production

    Scoped per project

    Security, evaluations, observability and rollout. The prototype becomes a system your team can run and your auditors can read.

    What you get
    Security, evaluations, observability and rollout, usually inside your own cloud subscription, so the keys, the data and the logs stay yours.
    How it is priced
    Scoped and quoted after the prototype, once what it takes is known.

  4. 04Operate

    Ongoing

    Monitoring, model and cost updates, and the next capability, with the same engineers who built it.

    What you get
    Monitoring, model and cost updates, and the next capability, with the same engineers who built the system.
    How it is priced
    Agreed with you once the system is in production.

Taking over existing work

Already have a prototype? We assess it first, keep what holds up, and replace what will not survive production, which is usually grounding, data handling and observability.

Where we work

We work remotely with clients across Canada, the United States, the Middle East and beyond, and in person across the Greater Toronto Area.

The first stage on its own is our AI strategy and readiness assessment. The terms used across the site are defined in the glossary.

Start with a conversation

Tell us what you are trying to do. We will come back with an honest take on what AI can do for it and what it would take to build.

Book an AI consultation