Summary
Taras Lishchenko is a Lead Computer Vision Engineer with over 10 years of experience building and optimizing real-world vision systems, from distracted-driver detection to warehouse robot navigation. He specializes in model performance optimization for edge platforms (TensorRT, ONNX, Jetson) and end-to-end product delivery, including MLOps pipelines and hardware-accelerated video processing. Taras combines deep research skills—implementing state-of-the-art detection and pose models—with pragmatic engineering (GStreamer plugins, real-time QR navigation at 120 fps on Raspberry Pi). He has led cross-functional teams and productionized models under imbalanced data and harsh environmental conditions, emphasizing reproducibility with DVC and MLFlow. An active technologist and blogger (lifestyletransfer.com), he is driven by automating tasks so people can focus on creativity, and he maintains hands-on fluency across Python, C++, ROS, and embedded platforms. Based in Vinnytsia, Ukraine, he brings a rare blend of low-level optimization and high-level product thinking to edge AI challenges.
10 years of coding experience
9 years of employment as a software developer
Master’s Degree, Computer Science, Master’s Degree, Computer Science at Vinnytsia National Technical University
English, Ukrainian, Russian, Spanish