Machine Learning Engineer - Computer Vision
Posted Updated
Description
Responsibilities:
- Lead the design, development, and optimisation of computer vision models for biometric applications, including face detection, recognition, facial attributes, and image quality assessment.
- Conduct fairness analysis and benchmark biometric models across diverse datasets, demographics, and real-world operating conditions.
- Design, train, and optimise machine learning models using PyTorch, TensorFlow, and/or JAX.
- Own end-to-end ML pipelines, covering data ingestion, preprocessing, model training, evaluation, and deployment.
- Build and maintain automated data ingestion and processing workflows using Apache Airflow.
- Curate balanced, high-quality training datasets and generate synthetic data to address data diversity gaps and improve model performance.
- Optimise models for low-latency inference using techniques such as quantisation, knowledge distillation, TensorRT, and ONNX.
- Deploy and scale ML solutions on AWS, ensuring production readiness, reliability, and performance.
- Mentor ML engineers, conduct code and design reviews, and establish engineering best practices across the Computer Vision team.
Requirements:
- 5+ years of industry experience in Machine Learning, with strong expertise in Computer Vision and biometrics.
- Hands-on experience developing and deploying biometric models for face detection, recognition, facial attributes, or image quality assessment.
- Practical experience in algorithmic fairness, bias analysis, and mitigating disparate impact in computer vision models.
- Strong proficiency in Python and ML/CV libraries such as PyTorch, TensorFlow, OpenCV, and Pillow.
- Experience designing and managing end-to-end ML pipelines, with working knowledge of workflow orchestration tools such as Apache Airflow.
- Hands-on experience with multi-GPU training and scaling ML workloads in distributed environments.
- Experience deploying and managing ML workloads on AWS services such as SageMaker, EC2, and EKS.
- Knowledge of model optimisation and inference acceleration techniques, including quantisation, distillation, TensorRT, and ONNX.
- Strong software engineering skills, with the ability to write clean, modular, maintainable, and production-ready code.
- Experience mentoring engineers and contributing to technical decision-making through code reviews and system design discussions.
Skills
Sign in to read and join the discussion.