Computer Vision and Machine Learning Engineer

15 hours ago

Toronto, ON, Canada Skymarkaerospace Full-time €90,000 - €130,000 Contract
Computer Vision & Machine Learning Engineer Careers // Position 04 Computer Vision & Machine Learning Engineer UAV Design and Development Status Open Skymark Aerospace develops uncrewed aircraft systems,with projects spanning aircraft platforms, propulsionintegration, and onboard sensing technologies. Our engineering work brings mechanical, electrical,and software disciplines together to develop and testpractical UAV systems. We are seeking a Computer Vision and Machine Learning Engineer to develop visual detection, classification, and tracking capabilities for UAV projects. You will work with software engineers, hardware specialists, technicians, and pilots to take models from experimentation through evaluation and deploymenton onboard computing hardware. The role combines model development with practical software integration. You will assess performance using representative imagery and flight-test data, investigate failurecases, and balance accuracy with processing speed, memory, and power constraints.
- 01 Develop, train, and evaluate computer vision models and algorithms for object detection, classification, and tracking in images and video.
- 02 Build and curate datasets, define annotation guidance, review label quality, and maintain versioned training, validation, and test sets.
- 03 Define evaluation methods and acceptance criteria with the project team, including accuracy, false detections, tracking performance, and processing latency.
- 04 Analyse failure cases across lighting, backgrounds, object sizes, occlusion, and camera motion; identify data and model improvements.
- 05 Integrate camera inputs, preprocessing, inference, and model outputs with onboard software and computing hardware.
- 06 Profile and optimise inference speed, memory use, and resource demand, verifying the effect of changes on model performance.
- 07 Maintain reproducible experiments, source code, model versions, configurations, and evaluation results so comparisons and releases can be repeated.
- 08 Support bench and flight-test data collection with pilots and test engineers, compare field results with laboratory benchmarks, and perform regression checks.
- 09 Document model assumptions, dataset limitations, operating constraints, and validation findings to support engineering decisions.
- 01 Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Applied Mathematics, or a related discipline, or equivalent relevant experience.
- 02 Demonstrated experience developing and evaluating computer vision or machine learning solutions using real image or video data.
- 03 Strong Python programming skills and experience with a deep learning framework such as PyTorch or TensorFlow.
- 04 Practical knowledge of image processing and detection, classification, or tracking methods, using tools such as OpenCV.
- 05 Understanding of dataset preparation, appropriate evaluation metrics, overfitting, and prevention of data leakage between training and testing.
- 06 Experience with Linux, source control, debugging, and packaging software or models for repeatable execution.
- 07 Ability to analyse experimental results, explain model limitations, document technical decisions, and collaborate across engineering disciplines.
- 01 UAVs, robotics, or other systems that process visual data in changing outdoor conditions.
- 02 Deployment on edge computing hardware such as NVIDIA Jetson, including inference optimisation with tools such as ONNX Runtime or TensorRT.
- 03 C++ development, camera integration, video pipelines, or performance profiling of deployed applications
- 04 Camera calibration, image geometry, small-object detection, or tracking from moving cameras.
- 05 Experiment tracking, model and dataset versioning, automated evaluation, or repeatable model deployment workflows.
- 06 Coordinating data collection and annotation, including identifying gaps that limit model performance. You will help turn experimental perception models into onboard capabilities with measured performance and clearly understood limitations. Your work will connect dataset quality and model decisions with results from actual hardware and field testing, guiding the team's next development priorities.