What we build, and what you receive.

Factories, fleets, queues and teams rarely behave like the spreadsheet average. The cheapest place to discover that is inside a model: a small virtual world that can be tested, challenged and improved before money moves. We build discrete-event, agent-based and physics-based simulations of queues, assets, logistics, resources and uncertainty.

• Where does the design break under real movement: queues, bottlenecks, capacity?
• What does the transition cost, down to the unit economics?
• Which operational policy still holds when the inputs are uncertain?

A working model with its documentation, the runs and evidence behind every claim it makes, and a handover so your team can run and extend it. You own all of it.

Decision tools, not products. Custom software, decision-support dashboards, scenario explorers and internal applications that replace brittle spreadsheets: engineered to a brief and owned by the client. We build the thing, not a slide deck about the thing, and we make sure it is sustainable.

• Which spreadsheet has quietly become critical infrastructure?
• What would the planning cycle look like as a tool the team actually uses?
• Can the model we built become something operators run every week?

Full ownership at handover: source, documentation, and a team that knows how to run it.

Statistical and machine-learning models, and applied LLM tooling, built into real engineering workflows rather than trend decks. Where a simple model answers the question, we say so and build the simple model: no black box if a simple model will do.

• Where does AI genuinely change this workflow, and where is it theatre?
• Can the system's output be checked by the people who rely on it?
• What does it take to run this inside our own environment?

A scoped, working system, the evidence from its evaluation, and the documentation to hold it to account.

Independent advice for boards, programmes and technical leaders on where AI changes the answer, and where it does not. Roadmaps, audits and decision-support architecture for leaders deciding what to build, what to buy, and what to leave alone.

A 12-to-24-month roadmap sequenced by value, risk and team readiness, with build, buy or partner scenarios. Audit and assurance. Technical due diligence on investment theses. Briefings and senior sessions. A standing advisory retainer where useful.

We are not strategists who would not know how to ship the thing we are recommending. Every roadmap we produce we could, in principle, deliver ourselves.

Training, briefings and embedded coaching that lift teams from AI-curious to AI-productive: practical, and grounded in the engineering work they are actually doing. We deliberately avoid generic training. Sessions are multi-day and calibrated to a specific team, a specific problem and a specific dataset.

Curricula and learning pathways. Embedded delivery alongside your engineers. Senior leadership briefings. Knowledge transfer designed so the practice survives our exit, and, where it helps, an internal community of practice.