Istanbul, Türkiye
I am a Software Engineer specialized in artificial intelligence, computer vision, and reinforcement learning, with a focus on health technologies and defense industry applications. I combine my internship experience in defense technologies at HAVELSAN with the innovative medical diagnostic system I developed as my capstone project. My goal is to apply my deep technical expertise to R&D processes, transforming theoretical models into practical, real-world solutions that deliver measurable impact.
As part of the TEKNOFEST 2025 – Technology for the Benefit of Humanity Competition, we formed the TEKNOŞİFA team to develop an AI-based decision support system aimed at accelerating and objectifying the diagnosis of acute appendicitis in pediatric patients using ultrasound images. Using the Regensburg Pediatric Appendicitis dataset, we automatically measured the appendix diameter with YOLO-assisted U-Net models (ResNet34, EfficientNet), and combined this feature with clinical data for classification using XGBoost. Instead of a traditional IoU metric, we designed a custom validation function based on diameter deviation, and deployed the system via a Docker-based web interface for clinicians. We also continued to develop the project during the 4-month inzva AI Projects program, where we presented our results to an audience of academics and industry professionals.
During my internship at HAVELSAN, I would like to express my sincere gratitude to my mentor Mr. Tunahan TUNA, as well as Mr. Fırat BOZKAYA and everyone at the HAVELSAN family, for generously sharing their knowledge and experience with me and helping me gain new perspectives.