Division 03 / Second Spark Intelligence
Software & AI
Product engineering and applied AI consultancy. The same team that built a payment rail and a battery diagnostics platform is available to build whatever you are working on.
Mobile · Web · Backend & cloud · Applied ML · AI integration
Position
We build for ourselves first.
Two of our three divisions are products we designed, built and now run. That is an unusual thing for a consultancy to be able to say, and it changes how we work: we have had to live with our own architecture decisions, our own estimates and our own shortcuts.
So we tend to be blunt about scope, conservative about complexity, and specific about what a thing will cost to keep running once it exists.
Engineering — 04
Software development.
Mobile applications
Cross-platform iOS and Android products, from first prototype through store release and the versions after it.
Web platforms & dashboards
Customer-facing products and the internal tools behind them — reporting, operations consoles, admin systems.
Backend, APIs & cloud
Services, data stores, authentication and the infrastructure to run them, sized for what you actually have rather than what a deck imagined.
Systems integration
Making software that was never designed to talk to each other do exactly that — payment networks, hardware, third-party APIs, legacy systems.
Intelligence — 04
AI consultancy.
Predictive modelling
Forecasting from your own operational data. The state-of-health models behind our battery work are the same discipline applied to a different domain.
Data pipelines & instrumentation
Most AI problems are data problems first. We build the collection, cleaning and storage layer before anything is trained on it.
AI feature integration
Language models and ML put into real products — assistants, extraction, classification, search — with the evaluation to know whether they are working.
Strategy & technical review
Where AI genuinely helps, where it does not, what it will cost to run, and an honest read on someone else’s technical claims.
How we work
Four stages, no surprises.
Scope
We start with the problem rather than the technology, and say plainly if the thing you asked for is not the thing you need. Fixed, written scope before anyone writes code.
Prototype
Something running early — narrow, real, and usable enough to argue with. Decisions get cheaper the sooner they meet reality.
Build
Short cycles, working software at the end of each one, and nothing kept behind a curtain until a launch date.
Hand over or operate
Documented and handed to your team, or run and maintained by ours. Either way you own the code.
In-house work
What we have built for ourselves.
| Product | What it is | Our scope |
|---|---|---|
| MiKi | Consumer payment wallet for local and cross-border transfers | Mobile app, backend services, payment integration |
| Battery Passport | Cell-level diagnostics and certification platform | Test tooling, data pipeline, state-of-health modelling |
| Fleet monitoring | Continuous battery health tracking and alerting | Telemetry ingest, predictive models, reporting |