Pocket Paralegal App
A mobile app that uses on-device AI to scan legal contracts and identify unfavorable terms, hidden fees, and risky clauses with precise visual highlighting of problematic text.
I'm a Computer Science graduate from The University of Sheffield with First Class Honours, passionate about building solutions at the intersection of machine learning and software development.
My interests span across a variety of domains including artificial intelligence, computer vision, and full-stack development. I enjoy applying cutting-edge techniques to solve real-world problems, and tackling complex challenges.
When I'm not programming, you'll find me bouldering, exploring astronomy, or tinkering with microelectronics. I'm fluent in both Turkish and English.
Sep 2021 — Jul 2025
First Class Honours
Oct 2022 — Mar 2023
A mobile app that uses on-device AI to scan legal contracts and identify unfavorable terms, hidden fees, and risky clauses with precise visual highlighting of problematic text.
Achieved 74% classification accuracy by developing deep learning models (CNNs, RNNs, Transformers) to classify harmful brain activity from EEG signals.
A multi agent simulated trading environment for automated trading strategies using reinforcement learning.
Cross-platform mobile app for meal recommendation and preparation using React Native, Expo, FastAPI, and PostgreSQL with Docker containerization.
A 3 & 5 dominos environment to train an intelligent agent that plays the game of dominoes using reinforcement learning.
A web application that enables companies to monitor employee mental wellbeing through customisable surveys with role-based access and analytics dashboards.
A repository for training various simple image generation models to generate 16x16 pixel art images.
Analysis and visualisation of embedding spaces for image generation machine learning models.