2 months ago
Montréal, CanadaSenior
Responsibilities
- Design and develop computer vision models and pipelines for biometrics use cases including face detection, quality assessment, and recognition.
- Benchmark models across datasets and operating conditions.
- Train and optimize models using PyTorch, TensorFlow, and/or JAX.
- Contribute to end-to-end ML pipelines from data ingestion through deployment, including automated Airflow pipelines for data ingestion and cleaning.
- Optimize models for low-latency inference using quantization, distillation, TensorRT, and ONNX, and support deployment on AWS.
- Collaborate with and mentor ML engineers and contribute to technical best practices across the Computer Vision team.
Requirements
- 5+ years of industry experience in machine learning focused on computer vision.
- Strong computer vision fundamentals, including experience with areas such as image classification, object detection, segmentation, image quality, or generative models.
- Awareness of algorithmic bias and willingness to learn domain-specific fairness practices.
- Strong proficiency in Python and related libraries such as Pillow, OpenCV, and PyTorch, with experience writing clean, modular, production-ready code.
- Experience designing or contributing to ML pipelines and familiarity with orchestration tools such as Airflow.
- Experience with GPU-based training and deploying ML services on AWS.
- Preferred qualifications include research publications in CVPR, ICCV, or ECCV; familiarity with vector databases and ANN search; experience with GANs or diffusion models for data augmentation; CoreML, LiteRT, or TFLite experience; and familiarity with privacy, security, and compliance considerations in sensitive ML applications.
Categories
