Things we build, and what breaks.

Interactive demonstrations of the workflows we build, plus the experiments behind them. Everything here runs on synthetic data.

Engineering experiments

The work behind the demonstrations.

Experiments built to answer a specific technical question. We publish the approach, the architecture and the failure modes — and the results once they have been measured.

On results

Experiments marked in progress have a defined method and architecture but have not been measured yet. We publish numbers once they exist, not before — an estimated benchmark is not a benchmark.

Recognise one of these problems?

If any of this looks like something happening in your business, tell us how it works today. We will tell you whether it is worth measuring.

20-minute conversation · No ERP integration required