ML Engineer

Requirements:

• Containerisation (Docker): Packaging vision environments with GPU drivers (CUDA/cuDNN).
• Azure Machine Learning: Using Azure ML for model versioning and automated retraining.
• Model Optimisation: Using TensorRT, ONNX, or OpenVINO to speed up inference.
• Edge Deployment: Experience with NVIDIA Jetson, Raspberry Pi, or Azure Percept.
• Data Annotation Ops: Managing tools like CVAT, LabelImg, or Roboflow for high-quality labelling.
• Video Streaming: Knowledge of RTSP, WebRTC, and FFmpeg for processing live video feeds.
• GPU Management: Understanding how to allocate and optimise VRAM for large batch training.
• MLOps: Setting up pipelines for "data-centric AI" (improving the data to improve the model).

Skills

Docker, ML, MLOps, Machine Learning