Georgy Makarov

Education

Peter the Great St. Petersburg Polytechnic University (SPbPU), Institute of Machinery, Materials, and Transport

Bachelor’s program: Mechatronics and Robotics, Autonomous Robots (2021 – 2025)

Higher School of Economics — St. Petersburg, School of Computer Science, Physics and Technology

Master’s program: Machine Learning and Data Analysis (2025 – present)

Participant of combined Master's-PhD track

Experience

Software Engineering Intern — Geoscan Group (Feb 2023 – May 2024)

Student Design Bureau

• Developed software in Python and C++, primarily in computer vision and image processing
• Created AI-based automation tools for Agisoft Metashape photogrammetry software
• Developed video transmission systems for gimbal-based drone cameras
• Configured, tested, and integrated electronic modules into the company’s software ecosystem

Contributed to organizing the nationwide competition “Personnel for the Digital Industry” (Cyberdrome 2023)

Junior ML Engineer — Geoscan Group (Jun 2024 – Feb 2025)

Software Department

• Developed and trained ML models (CNNs, transformers) for computer vision tasks for geospatial data analysis
• Maintained and upgraded the company’s MLOps infrastructure
• Integrated ML models into Geoscan projects and deployed them through online services
• Designed algorithms for UAV aerial survey data processing (sensor fusion, point-cloud georeferencing, image matching)

ML Engineer — Geoscan Group (Mar 2025 – present)

Software Department

• Authored and coauthored scientific papers and reports
• Conducted ML research for geospatial analytics within Geoscan projects
• Developed and trained ML models (CNNs, transformers) for computer vision tasks for geospatial data analysis
• Maintained and upgraded the company’s MLOps infrastructure
• Integrated ML models into Geoscan projects and deployed them through online services
• Designed algorithms for UAV aerial survey data processing (point-cloud clustering, real-time photo georeferencing)

Best student report at the 27th Conference of Young Scientists “Navigation and Motion Control” — “Detection of Road Obstacles on a Digital Terrain Model”

Publications

• Under peer review: “GenAI in Digital Avatar Synthesis: A comprehensive review”
• Under peer review: “Effective Sparse Data Processing in Vision Transformers”
• 27th Conference of Young Scientists “Navigation and Motion Control” (2025): “Road Obstacle Detection on Digital Elevation Model” (abstract)

Skills

• Python (OOP, multithreading, async programming)
• Discriminative CV: CNNs, transformers (ViT, DETR, SAM), SSMs, distillation, evaluation
• Generative CV: autoencoders (AE, VAE/VQ-VAE), GANs, diffusion models (DDPM/DDIM, LMD), flow matching, adapters & personalization, distillation, evaluation
• Theoretical base on ML (classical ML, NLP, model pruning & quantization, CUDA)
• Frameworks: PyTorch, Transformers, Diffusers, OpenCV, Scikit-Learn, Pandas
• Linux, Git, Jupyter, coding standards
• Languages: Russian — native, English — advanced

Personal Qualities

Motivated to learn and grow; responsible, stress-resistant, disciplined; strong team player. Aimed at developing academically in computer science and machine learning through journal publications, conference reports, and further education.