описание
TargetAI specializes in video analytics solutions.
задачи
- Design and develop computer vision and video analytics models for object detection, tracking, and recognition
- Work with video pipelines, including RTSP/HTTP, decoding, frame sampling, and batching
- Adapt and customize models for real-world conditions, including poor lighting, different camera angles, noise, compression, and FPS drops
- Fine-tune and retrain models for specific customer scenarios
- Optimize inference for CPU
- Reduce latency and resource consumption
- Balance accuracy and performance
- Integrate ML models into production backend services, edge devices, and on-prem installations
- Monitor model quality, including precision/recall and data drift
- Prepare and analyze datasets through annotation, validation, and data augmentation
- Build training and testing pipelines
- Analyze model errors and quality degradation
требования
- 4+ Years of experience in Computer Vision/Video Analytics
- Strong understanding of CNN, YOLO, and SSD
- Experience with PyTorch or TensorFlow, OpenCV, and ONNX
- Understanding of video formats and streams, including RTSP, H.264/H.265 codecs, FPS, bitrate, and latency
- Experience optimizing CPU-only inference
- Experience working with Linux
- Deep understanding of OpenVINO
- Knowledge of edge device logic
- C++ skills for performance-critical parts
- Experience with Kafka and gRPC/REST
- MLOps skills, including MLflow, model versioning, and monitoring
- Experience with large-scale CCTV projects
условия
- Условий в вакансии нет
навыки
компания
IT-компания, которая создает продукты для упрощения практического применения генеративного ИИ. Команда из более чем 50 экспертов в области голосовых технологий и ИИ разрабатывает гибкие и инновационные решения. Компания сотрудничает с крупными клиентами, такими как Билайн, Яндекс, Авито и Газпром.
Подробнее о компании →