I build systems that have to prove what they did.
Pay-transparency reporting, AI Act release gates, transport equity. Regulated domains where the answer is worth little without the evidence behind it — so the audit trail is the product, not a feature bolted on at the end.
5
systems live in production
8
packages published on PyPI
1
working paper, with DOI
27
EU states in the panel
Selected work
Production systems
WorkforceGuard AI
LiveTurns EU labour-market data and a company's own payroll into a pay-gap position it can defend to a regulator.
Aequitas
LiveAnswers one question the same way in every country: who here is underserved by public transport, and by how much?
EU AI Assurance OS
LiveA fail-closed release gate for AI shipped into the EU: a pinned legal corpus, a human proposal queue, and an Evgraph 0.1.2 scan that must match the evidence pack.
MaSoVa Restaurant OS
LiveA full restaurant operating system: ordering, kitchen, delivery, payments, analytics and compliance across multiple stores and tax jurisdictions.
Open-source libraries
Evgraph
PyPITurns Model Cards, approvals, and deployments into one Evidence Graph. EU AI Assurance OS pins evgraph-cli 0.1.2 so the evidence pack and the scan share the same gap.
Moveq
LivePython libraries that turn service and demographic data into standard inequality measures — and keep cross-country methods honest.
Research
Why Tight Labour Markets Do Not Close Gender Pay Gaps: Evidence from a 27-Country Eurostat Panel
Tight labour markets are supposed to close gender pay gaps through competition for workers. Across 27 member states and six years, the association runs the other way. The panel behind that finding is the same warehouse that serves WorkforceGuard in production.
The finding, the method, and the data