Leading Computer Vision Engineer

Main requirements

1. 4+ years of experience in Computer Vision/Video Analytics.2. Excellent understanding of the work:- CNN, YOLO, SSD- Face recognition pipelines- OCR/ALPR3. Extensive experience with: PyTorch or TensorFlow, OpenCV, ONNX.4. Understanding how video formats and streams work: RTSP, codecs (H.264/H.265), FPS, bitrate, latency.5. Experience optimizing for CPU-only inference.6. Experience working with Linux.7. Deep understanding:- OpenVINO- Logic for working with edge devices- C++ for performance-critical parts- Kafka- gRPC/REST8. MLOps skills: MLflow, model versioning, monitoring.9. Experience working with large-scale CCTV projects.

What you will do

1. Design and development of CV/Video Analytics models: object detection (person, vehicle, face, etc.), tracking, recognition (faces, license plates, attributes).2. Working with video pipelines:- RTSP/HTTP decoding, frame sampling, batching3. Adaptation and customization of models to real conditions: poor lighting, different camera angles, noise, compression, FPS drops.4. Fine-tuning and further training of models for specific customer scenarios.5. Optimization of inference for the CPU.6. Reducing latency and resource consumption.7. Balancing accuracy vs performance.8. Integration of ML models into production: backend services, edge devices, on-prem installations.9. Monitoring the quality of models:- precision/recall- drift data 10. Preparation and analysis of datasets: markup, validation, data augmentation.11. Building training and testing pipelines.12. Analysis of model errors and quality degradation.

What we offer

- Coverage of medical services;- Compensation for sports activities;- Access to an educational platform with mini-courses and complete programs in various fields.