Пожалуйста, войдите, чтобы скачать резюме.
Experience level:
3-5 years
Education level:
Languages:
Russian: Native
English: B1
Computer Vision Engineer with 5 years of experience in this sphere. Developed different parts of systems. Worked with different tasks and technologies. I love to learn something new. Have a master degree and can decompose tasks to small pieces. Love different cultures and is open in communication. Good team worker. Try to listen everybody. Looking for a relocation to Serbia.
Developed educational materials and laboratory guidelines for Data Base Architecture.
Conducted online lectures and technical workshops for students.
Evaluated student projects and provided technical feedback.
Responsible for data gathering, training and fine tuning models, developing software for using
CV models for a business tasks.
Developed Face Recognition System for Retail by using Python, DeepFace, Apache TVM (for
deploying on Mali GPU of a NanoPi microcomputer), PostgreSQL and Grafana.
Updated Smart Scales System for Retail – changed object classification model to a
segmentation model for better accuracy. Used Python, Pytorch, MobileNet.
Developed Smart Lock System – reading key-cards for opening doors with syncronization of
allowed keys with business owner’s systems. Used C++, PostrgreSQL, NFC-cards reader’s
SDK.
Developed Smart Greenhouse IoT Platform. Built an HTTP API server to orchestrate Arduino-
based actuators and USB-cameras. Used Flask, Python.
Responsible for data gathering, training and fine tuning models, developing software for using
CV models for a business tasks.
Took over ownership of Smart Scales for Retail system – fixing bugs, retraining models,
troubleshooting, adding small features. Used Python, Pytorch, CVAT SDK, Docker, Grafana.
Developed Obstacle Detection System for Industrial Robotics. A safety feature for an automatic
forklift in extreme conditions (occluded with plastic wrap cameras). Developed a YOLO-based
segmentation model capable of detecting obstacles through film-covered lenses. Performed
model export and optimization for RKNN (Rockchip NPU) to run on Orange Pi edge boards.
Integrated the vision system with the ROS (Robot Operating System) ecosystem for real-time
robotic control. Used: YOLO, RKNN, ROS, Python, Orange Pi.
Developed Retail Inventory Monitoring System. Trained and deployed YOLO segmentation
model for detecting transport boxes in 100+ stores with information from IP-cameras. Designed
the architecture of a PostgreSQL data base for the project. Developed a Telegram-based alerting
system for real-time inventory notifications. Used: Ultralytics/YOLO, Python, Aiogram,
PostgreSQL.
Mantained a Warehouse Inventory Robot. Fixed bugs, increased reliability, documented code.
Used C++, Gstreamer, Docker.
• Core: • Python, Pytorch, OpenCV, YOLO, NumPy, PostgreSQL, Docker, Git, Grafana. • Familiar with: C++, Apache TVM, ROS, RKNN-toolkit, Gstreamer, HTML/CSS/JavaScript. • Soft Skills: Teamwork, Adaptability, Problem-solving, Cross-cultural communication.