Skip to content

The goal changed. We didn't.

Applied AI, backend and security, optimisation and modelling — three disciplines, one collective, and every project published with the numbers it actually produced.

Projects

Football Squad Optimizer

A decision, and what it rests on.

A Fantasy Premier League squad is a constrained optimisation problem, so SquadOpt solves it as one: a £100m budget, the squad rules and a projection for every player go in, and CP-SAT returns not a good squad but a proof that no better one exists under those constraints. Every gameweek it publishes what it projected beside what actually happened — including the weeks the projection lost.

  • Python
  • OR-Tools CP-SAT
  • ML
  • React
ML models promoted to production
0
Football Squad Optimizer screenshot

Who we are

We met at Antalya Anadolu Lisesi, one of Turkey's most prestigious high schools, and grew close studying for the university entrance exam — three friends held together by a shared goal. The goal has changed; the shape has not. We are here to do the work as well as it can be done, to catch each other's mistakes before they ship, to teach each other what we know, and to enjoy it.

  • An observation becomes an issue, with evidence.
  • A handed-over plan is checked against the code, not believed.
  • Every change says what it did not change.
  • Thresholds are written before the work, never fitted to it.
  • A claim is not a measurement.

Team

  • İbrahim Ersan Özdemir

    İbrahim Ersan Özdemir

    Full-stack & applied AI

    Computer engineering graduate. React and React Native on the surface; ANNs, LLMs and retrieval pipelines underneath RAG with local inference, computer vision, and machine learning on real datasets.

  • Tunay Aslan

    Tunay Aslan

    Backend & security engineering

    Computer engineering student at Boğaziçi. Builds systems that watch other systems: a secret-scanning pipeline across an organisation's repositories, and a routing engine over a hundred thousand hosts.

  • Ertuğrul Soydal

    Ertuğrul Soydal

    Optimisation & modelling

    Computer science and industrial engineering, double major, mathematics minor at Sabancı. Optimisation and modelling: Gurobi, statistical analysis, and the maths under the models. Also taught calculus.