SERVICE / APPLICATION DEVELOPMENT
Software that works where you actually use it.
Most software is designed for an office with fibre. We design for the job site, the production line, the vehicle, the warehouse floor, and the places where a spinner just means the app is broken. Whatever the constraint — offline-first, low-spec devices, a team that doesn't want to learn new software — we build for the real conditions, not the demo conditions.
What we build
- Progressive Web Apps (PWA) — installable, offline-capable, no app store gatekeeping
- Internal business tools built rapidly on low-code platforms (e.g. AppSheet), for teams that need something running this month, not this year
- Custom web applications and dashboards (audit-ready dashboards, track-and-trace applications)
- Industrial software — IoT integration and PLC-connected applications
- Mobile-first field data capture tools (inspections, logging, compliance forms)
- Integrations between the tools you already use and the ones you're about to need
Under the hood
Built primarily on React, Vite, and Tailwind CSS on the frontend — a stack chosen for speed, small bundle sizes, and long-term maintainability, not because it's trendy. Where the job calls for it, that extends to Python and FastAPI with PostgreSQL on the backend, offline-first storage and service workers for places without reliable signal, and native mobile via Capacitor when a PWA isn't enough. We build the AI-assisted and automated-analysis tooling ourselves too, rather than just wiring up someone else's platform.
If "but will it work when the signal drops" is a real question for your team, that's the conversation to have with us.
AI systems architecture, built for what happens after the demo. We design and govern AI systems for environments where things go wrong quietly — a sensor drifts, a model degrades, a decision gets made without anyone checking it. Our background is in the infrastructure layer: industrial IoT, radio networks, application systems built to survive the field. That's the vantage point we bring to AI architecture — not lab conditions, but production conditions, where trust in a system's output has to be engineered in, not assumed. If you're layering intelligence on top of operational infrastructure, we build it so the humans accountable for the outcome can actually stand behind it.