computer vision engineer for waste imagery

Sortrace develops a platform for analyzing waste imagery using computer vision. The platform is already in production.

• Develop, train, and optimize deep learning models for image analysis
• Evaluate and extend existing model architectures
• Improve model accuracy, performance, and robustness
• Process, augment, and evaluate image datasets
• Experiment with architectures and training approaches
• Solve technical problems and iterate on production computer vision systems

• At least 3 years of hands-on computer vision and model development, or equivalent documented experience through professional work, open source, or serious personal projects
• Experience building, training, evaluating, and optimizing deep learning models for images
• Solid understanding of computer vision architectures, such as CNNs and Vision Transformers
• Strong Python skills
• Practical experience with PyTorch or TensorFlow
• Experience preprocessing, augmenting, and evaluating real-world image datasets
• Ability to adapt and extend existing models to meet clear technical requirements
• Curious, solution-oriented, and willing to investigate technical problems in depth
• Keeps up with computer vision and deep learning
• Able to work independently and in a structured way, and collaborate on technical decisions
• Будет плюсом: Edge AI and running models on edge or embedded devices, optimization for limited compute including quantization and inference performance, real-time image or video analysis

• Production computer vision work with real impact
• Work on an existing platform that is already running and improve what ships
• Opportunity to influence architecture and methods
• Work on challenging problems where precision, robustness, and performance matter