🎯 Computer Vision Engineer | 🧠 Multi-Object Tracking Researcher | 🧪 Contributor @ BoxMOT

I'm a Computer Vision Engineer with 2+ years of hands-on experience in designing and deploying deep learning models for real-world vision problems. My primary focus lies in multi-object tracking (MOT) and object detection, where I have experience across the full pipeline — from data collection and annotation to model training, optimization, and production deployment.

My recent work includes:

• Creating own Tracker with independent Re-Identification, achieving SOTA-1 results on MOT17
• Contributing to the BoxMOT library (added the SOTA BoostTrack tracker, made metrics of many trackers reproducible)
• Real-time inference optimizations using TensorRT, ONNX, and Triton Inference Server
• Model training with PyTorch, Torch Lightning, and distributed computing with Ray