Skip to content
AD

Alexander Dolgushev

@LoaferCH

Computer Vision Engineer

CV Engineer with 5+ years of experience in production-level detection, segmentation and classification. Eager to find great opportunities in computer vision field.

Career

Experience: 6 years

Experience

Computer Vision Developer

PTF-Lab

Jan 2025 — Present

Develops real-time sports segmentation and synthetic-data pipelines, and converts models for deployment.

• Developed and optimized real-time semantic segmentation models for sports events for advertisement replacement pipelines • Converted deep learning models to ONNX and TensorRT for deployment workflows • Built synthetic data generation pipelines using Blender, Blender Python API, and Unreal Engine • Adapted scenes for rendering segmentation masks, animations, and synthetic datasets

Computer Vision Engineer / Team Lead

InnoGeoTech

Oct 2021 — Mar 2024

Built segmentation models and geospatial workflows for satellite imagery, while leading two junior developers.

• Developed semantic and instance segmentation models for geographical landmark detection using satellite imagery • Trained segmentation networks for rivers, forests, roads, swamps, and cloud detection using PyTorch • Performed geospatial data analysis using QGIS, Rasterio, and Python • Integrated trained neural networks into internal repositories and workflows using Git • Led and coordinated two junior developers

Computer Vision Engineer

RoadAR, Inc.

Jun 2019 — Oct 2021

Developed road-scene vision pipelines and optimized models for embedded inference.

• Developed computer vision pipelines for lane detection, object detection, tracking, and road incident analysis • Researched and evaluated deep learning architectures for embedded deployment • Converted models to RKNN and ONNX formats for inference optimization • Implemented camera calibration and vehicle lane positioning systems using OpenCV and Python • Built abandoned object detection pipelines using background subtraction and neural networks • Performed dataset collection, auto-labeling, annotation, and model training using PyTorch and MMDetection

Machine Learning Intern

Provectus

Jun 2018 — Jul 2018

Analyzed mobile-app usage data and correlations with demographic groups.

• Performed data analysis on mobile application usage datasets • Identified correlations between application usage patterns, gender, and age groups

Education

Bachelor of Computer Science, Data Science

Innopolis University

2015 — 2019

Skills

PythonC++PyTorchOpenCVMMDetectionOpenVINORKNNONNXTensorRTDockerCVATGitQGISRasterioBlender Python APIUnreal EngineYandex Toloka