b
buraksahin78

Burak S.

@buraksahin78
Turquie
Anglais, Turc
Certaines informations sont présentées en anglais.
À propos de moi
Hi, I'm Burak, a Computer Engineer specializing in computer vision and deep learning. I train custom models for object detection, image classification and segmentation using YOLO, PyTorch and modern architectures. I've built a real-time traffic sign detection system for an autonomous vehicle team competing at Teknofest, and developed deep learning models for post-disaster building damage assessment in my thesis. I focus on models that actually work on your data, with clear reports on performance and limitations. Feel free to message me before ordering to discuss your project.... Plus d’infos

Compétences

b
buraksahin78
Burak S.
hors ligne • 

Voir mes services

Vision par ordinateur
I will train a custom yolo object detection model on your dataset
De90 $US / projetPlus de détails

Portfolio

Expérience professionnelle

Dokuz_Eylul University

Software Lead – Autonomous Vehicle Team

Dokuz Eylul University • Temps partiel

Oct 2025 - Jul 2026 • 9 mos

Led the software department of DEOS, a volunteer university autonomous vehicle team competing in Teknofest Robotaxi. - Built a full traffic sign detection pipeline: a 97k-image dataset with 30 Turkish sign classes - Solved severe class imbalance with perceptual-hash deduplication, targeted augmentation and stratified splitting - Benchmarked 9 YOLO models (v8, v11, v12) on a held-out test set; best model reached 0.91 mAP@0.5 - Deployed the detector on a Raspberry Pi 5 with a Hailo-8 NPU, running in real time at ~86 FPS - Co-designed the vehicle's ROS2 (Jazzy) software architecture and integrated the Intel RealSense depth camera - Coordinated the team's software work, code reviews and Git workflow