EdgeBoost
A multi-output Gradient Boosting Decision Tree framework that uses random projections in its objective to scale to extreme label counts. Outperformed traditional GBDTs by 11% on multi-output benchmarks.
Machine Learning Research Engineer in Barcelona — game security, behavioral AI, and anti-cheat detection.
Machine Learning Research Engineer with 7+ years shipping models to production. I work on the RICOCHET anti-cheat team at Activision, building behavioral and vision-based cheat detection for Call of Duty. Before games: Databricks pipelines, real-time analytics, and experimentation.
A multi-output Gradient Boosting Decision Tree framework that uses random projections in its objective to scale to extreme label counts. Outperformed traditional GBDTs by 11% on multi-output benchmarks.
A clone of 1010! with a search-based AI built on Monte Carlo Tree Search and custom heuristics, written from scratch in Python.
An offline, cross-platform Cribbage card game built in Unity for iOS and Android. Sole developer for 2 years — gameplay, AI opponent, UI, and live ops.
A Python crawler and analyzer for yemeksepeti.com that scrapes restaurant menus and pricing for delivery-market analytics.
3D Printing, Game Development, Gaming, Hiking, Bouldering, Cafe Hopping
This is the plain-HTML version of the profile, published for readers, screen readers, and crawlers that do not run JavaScript.