Responsibilities:
• Design multi-agent systems and integrate LLMs with internal tools
• Develop backend APIs, RAG pipelines, and tool/function calling
• Build computer vision pipelines for detection, tracking, and re-identification
• Optimize real-time video and GPU inference
Requirements:
• Node.js/Express.js and Python
• LLM APIs, prompt engineering, RAG, tool calling, or agent frameworks
• Docker, Linux, and Git
• PyTorch and OpenCV
• Object detection and multi-object tracking (YOLO, ByteTrack/BoT-SORT)
• Person re-identification, vector search, and video streaming (RTSP)
Контакты работодателя доступны по кнопке «Откликнуться» после входа.