Sectors
Where the constraints are real
We are not sector-agnostic, and we are not sector-exclusive. These are the domains whose constraints we have worked under long enough to be useful from the first conversation.
01
Defence and national infrastructure
Systems that must hold their properties under adversarial conditions, and demonstrate that they do.
Requirements in this sector are dominated by assurance, sovereignty and the constraints of operating where commercial infrastructure cannot be assumed. Custom hardware is routine, component provenance is a design constraint rather than a procurement detail, and the evidence that a system behaves as specified is as much a deliverable as the system. We have taken sensing and autonomy systems from schematic to working prototype under these conditions — board design, firmware, edge inference and the fusion software above them — and we work within the classification and nationality constraints of the programme.
Recurring problems
- RF sensing, classification and localisation of uncooperative targets
- Edge inference held inside a deployable power and thermal budget
- Autonomy that degrades predictably when the link does
- Sensor fusion with honest confidence, and the assurance evidence for it
02
Capital markets
Where the difference between adequate and correct is measured in microseconds.
Trading infrastructure is unusual in that its performance characteristics are its function. We work on execution and market data paths, latency budgets held end to end rather than per component, and the measurement discipline required to know whether a change helped. Our work spans equities, fixed income, derivatives and digital assets.
Recurring problems
- Latency budgets and hot-path engineering
- Market data ingestion, normalisation and replay
- Signal research infrastructure and evaluation harnesses
- Pre-trade risk and position management
03
Financial services
Where a model that cannot be explained cannot be used, however well it performs.
Banking, asset management and insurance share a constraint that shapes every technical decision: the institution must be able to justify its systems to a regulator and its risk to a committee. We build quantitative and operational systems on that basis — inspectable, documented, and designed so that model risk is visible rather than embedded.
Recurring problems
- Model risk, validation and documentation
- Risk engines and portfolio analytics
- Structured instruments and their risk bases
- Operational automation with auditable decisions
04
Energy, utilities and resources
Where the system is physical, and a control decision has consequences in the world.
Generation, distribution, trading and processing assets combine long-lived physical plant with instrumentation of variable quality. The recurring problem is the gap between what an asset is doing and what its data says it is doing. We work on process control, instrumentation and the modelling required to close that gap, alongside the trading and risk systems built on top of it.
Recurring problems
- Process control and virtual instrumentation
- SCADA and industrial data architecture
- Techno-economic assessment of process change
- Energy trading, risk and carbon measurement
05
Industrial and advanced manufacturing
Where custom hardware is the starting point, not an escalation.
Manufacturing, materials and R&D operations reach the limits of commercial platforms early, and the work moves to instrumentation, embedded control and simulation. We design systems that span that boundary — from a single added sensor through to integrated control and digital twins of a physical process.
Recurring problems
- Embedded control and instrumentation
- Digital twins and process simulation
- Predictive maintenance grounded in physical models
- Integration across plant, edge and enterprise systems
Enquiries
Start with the problem
If you are weighing a system decision that is difficult to reverse, we are glad to look at it with you. An initial conversation carries no obligation and no fee.