The same team, pointed at your product.
We take on a small amount of client work each year, in the areas we already build in for ourselves. It funds the products; it is not what we are.
Mobile apps
Design, build, ship and keep shipping — store submissions, review rejections, staged rollouts and the update treadmill afterwards.
Our proof: SparkKing, iOS + AndroidAI & ML systems
Speech scoring, document understanding, retrieval, evaluation harnesses — and the human review paths that decide whether any of it is safe to deploy.
Our proof: syllable scoring · coding co-pilotData platforms
Ingestion from payment, commerce and analytics sources; warehouse modelling; attribution that survives an argument with the finance team.
Our proof: KlarSoftware engineering
Backends, APIs and web apps we intend to still be running in three years. Boring, observable and cheap to operate.
Our proof: everything aboveConsulting
Where AI genuinely earns its keep in your product, where it will just cost you money, and what order to build it in.
Our proof: our own roadmapTry the products first.
The three previews on the home page are the portfolio. If they feel like work you'd want on your own product, that's the conversation.