Case study · AI & Product
The Stefan Kuryłowicz Foundation runs national architecture and theory competitions. We designed and built the software that runs the judging — anonymous submissions, weighted scoring, live ranking — plus the public website around it, delivered end-to-end with an AI-assisted workflow.
The foundation’s competitions — Teoria (essays on architecture) and Praktyka (built and design work) — were judged the hard way: submissions collected over email, jurors scoring on spreadsheets, results tallied by hand. Two things were at stake in every edition: fairness — a juror should never know whose work they are scoring — and trust in the final ranking. Both are hard to guarantee with files and spreadsheets passing between people.
End to end. We brought the concept to the foundation, designed the entrant and juror experience, defined the data model and scoring rules, and delivered the working software using an AI-assisted development workflow — product, design and delivery under one roof. The same engagement covered the foundation’s public website.
Around the platform, a new public site for the foundation: its mission, the PRAKTYKA competition, calls for entries and a lead-generation flow for young architects — the same quiet, typographic design language, built to be edited by the foundation’s team.

Website — home: mission-led, with an open call to the current competition.

Website — competition page: rules, timeline and entry call.

Website — gallery of past editions and laureates.
Entrants submit through one clean form; jurors score anonymized work against clear criteria from anywhere; the office watches the ranking build itself and exports the result. What used to be an email-and-spreadsheet process is now an auditable, fair, self-tallying platform — delivered alongside the website that fronts it.