Computer Vision Engineer
Core
Architecting robust, real-time video ingestion pipelines and optimizing inference for ultra-low latency on Edge and Cloud environments to enable live Vision AI on video feeds.
Role type
Systems-First Computer Vision Engineer (Production/Deployment)
Builds
High-performance streaming pipelines for live video feeds
Domain
Computer Vision, Real-time Video Streaming, Edge/Cloud Infrastructure
Deliverable
production ML models
Required skills
Video streaming protocols (RTSP, WebRTC, FastRTC), FFmpeg, GStreamer, OpenCV, Python (multi-threaded/async), Docker, Geometry, Statistics
Preferred skills
NVIDIA TensorRT, DeepStream, Triton Inference Server, PyTorch/TensorFlow (inference), Kubernetes, SQL/NoSQL
Technologies
GStreamer, FFmpeg, WebRTC, NVIDIA Jetson, Docker, Kubernetes, PyTorch, TensorFlow, SQL, NoSQL
Responsibilities
Design and implement robust real-time video ingestion pipelines handling multiple RTSP streams; Integrate trained models into production pipelines ensuring stability and low memory usage; Optimize data processing pipelines for low-latency inference on Edge and Cloud; Build fault-tolerant systems that recover gracefully from camera or frame drops; Maintain and evolve proprietary vision frameworks with modular Python code; Diagnose system bottlenecks (CPU, GPU, Network) and implement architectural fixes.