Mechanical Engineering student at METU with a passion for coding. I build live web applications, computer vision tools, desktop GUIs, and automation software.
An intelligent computer vision assistant that analyzes game states in real-time to provide players with automated visual tracking and strategic insights.
A browser-based, privacy-first PDF toolkit. Merge, split, compress, convert, watermark, and edit PDF files client-side with zero data uploaded to external servers.
Computer Vision • Real-Time Analysis • Game Strategy
The Challenge: Build a system that can visually interpret a live Okey game board, recognize tiles in real-time, and provide strategic suggestions — all running directly in the browser without server-side processing.
The Solution: Okey Helper uses a custom computer vision pipeline to detect and classify game tiles from a live camera feed. The system processes each frame locally, identifies tile patterns, and calculates optimal moves using game theory algorithms.
Key Technical Decisions: Client-side processing was chosen to eliminate latency and ensure privacy. The tile recognition model was trained on a custom dataset of Okey tile images to achieve high accuracy across varying lighting conditions and camera angles.
Client-Side JavaScript • Privacy Architecture • PDF Processing
The Challenge: Most online PDF tools upload your files to remote servers for processing, creating privacy and security risks. The goal was to build a full-featured PDF toolkit that processes everything in the browser — no uploads, no servers, no data retention.
The Solution: MassForge PDF leverages pdf-lib and pdf.js to perform all operations (merge, split, compress, convert, watermark, edit) entirely in JavaScript. Files never leave the user's device, and all data is discarded when the tab is closed.
Key Technical Decisions: A modular architecture was implemented so each PDF operation is an independent module. This allows lazy-loading of features, keeping the initial bundle size small while supporting a full suite of PDF tools.
High-performance batch media crawler & downloader built with PyQt6. Features Playwright headless JS rendering, segmented downloading, and FFmpeg auto-muxing.
Native offline desktop software engineered with Python and CustomTkinter for bulk PDF manipulation and format conversion without cloud reliance.
Visual automation script leveraging Python, OpenCV, and Tesseract OCR for visual game recognition and anti-AFK kick protection.
A mod menu tool built for Thief Simulator 2, providing feature enhancements and customizable script utilities.
Custom mod menu software designed for Used Cars Simulator with interactive controls and game enhancements.
Yes! Both Okey Helper and MassForge PDF are completely free to use with no registration required. Simply visit the app URL and start using them immediately.
Absolutely safe. MassForge PDF processes all files 100% client-side in your browser. Your files are never uploaded to any server. When you close the browser tab, all data is completely discarded. Zero data retention, zero uploads.
My core stack includes Python, OpenCV, Tesseract OCR, CustomTkinter, PyQt6, Playwright, JavaScript (ES6+), HTML/CSS, and Git. I specialize in computer vision, desktop GUI applications, and browser-based web tools.
Yes, most of my desktop applications and tools are open source and available on my GitHub profile. Feel free to explore the source code, open issues, or contribute.
The best way to reach me is through my GitHub profile. You can open an issue on any repository or reach out directly. I guarantee a response within 24 hours.
I am Ataberk Susam, a Mechanical Engineering student at Middle East Technical University (METU) in Ankara, Turkey. Alongside my engineering studies, I focus on software development, specifically computer vision algorithms, desktop user interfaces, and web tools.
Core Stack & Tools: Python, OpenCV, Tesseract OCR, CustomTkinter, PyQt6, Playwright, JavaScript (ES6+), HTML/CSS, Git.
Have a question, want to collaborate, or just want to say hello? Reach out through any of the channels below.