Computer Engineering student at İzmir University of Economics (3rd year), focused on AI/ML and backend development. I like building tools that solve problems I actually have, and contributing fixes upstream when I run into bugs.
| Project | Description |
|---|---|
| Molt | Native macOS file converter built with Swift/AppKit, powered by the Filee engine. Finder "Convert with Molt" action, TR/EN UI, Apple Silicon & Intel DMGs released via GitHub Actions. |
| Self-Hosted WireGuard VPN | Personal VPN on Oracle Cloud: wg-easy in Docker, token-protected Flask API behind Caddy HTTPS, web dashboard and a Chrome extension for server status. |
| Lectoprint | Claude skill that turns lecture slides into exam-ready study PDFs with a Python/ReportLab pipeline: full notes, quick review or question bank, in English, Turkish or hybrid. |
| Social Media Downloader | Lightweight Manifest V3 Chrome extension for downloading Instagram stories and TikTok videos. |
| SportsManager | 5-person team project in JavaFX/Maven: design patterns, 13 screens, 27 unit tests, CI builds for macOS (.dmg) and Windows (.exe). |
- Filee: regular contributions to the cross-platform conversion engine behind Molt, e.g. JSON event streaming for the CLI
- Vorssaint: implemented an opt-in "Dim the picture" mode for external displays over DDC (#3837, merged)
- IEUStudia: study app for IEU students
- cobanov/instagram: Add Chrome extension wrapper, merged; added as a contributor
AI Intern @ Teknology House · Summer 2026
Computer vision: object detection and segmentation with YOLO.
Languages: Python · Swift · Java · JavaScript
Frameworks & Libraries: Ultralytics YOLO · PyTorch · Flask · ReportLab · AppKit · JavaFX
Platforms: macOS · Chrome Extensions (MV3) · Oracle Cloud · Linux
Tools: Docker · Caddy · WireGuard · GitHub Actions · Maven · Git


