Applied Machine Learning Scientist

at  Distran AG

WHO WE ARE

Distran is a fast-growing, award-winning, Swiss high technology company producing an extraordinary product: a sensor that literally sees the sounds. Our unique product is used by major global players in Oil and Gas, Chemical, Power and even Space Exploration to detect gas leaks before they become dangerous to either humans or the environment. With strong growth in 35+ countries, our company continues to expand thanks to our amazing people passionate about innovation, climate change, and protecting the environment.

OPTIONAL QUALIFICATIONS:

• Hands-on experience applying MLOps practices, including data versioning, experiment tracking, and model deployment.
• Familiarity with off-the-shelf deep learning models and their application in real-world scenarios.
• Experience in deploying ML models and familiarity with tools like ONNX to optimize model deployment on constrained compute architectures.
• Experience with data visualization tools for communicating insights and model performance.
• Understanding of hardware constraints and embedded systems relevant to machine learning applications.

Join Distran in shaping the future of ultrasonic imaging and leak detection with the world’s first ultrasound camera!
As an Applied Machine Learning Scientist specializing in quantitative modeling, you will develop, train, tune, and deploy advanced machine learning models that enhance the performance of our cutting-edge technology. You will lead key innovations in leak quantification and classification, depth estimation, scene understanding, tracking, and multimodal sensing. Your contributions will directly improve the accuracy, efficiency, and capability of our systems, driving the evolution of our ultrasound cameras and the future of leak detection.

Key Responsibilities:

• End-to-End Model Development: Design, implement, and optimize machine learning models for quantitative tasks in ultrasound imaging and leak detection. Navigate efficiently through a variety of machine learning, statistical, and deep learning techniques to identify the most suitable models and architectures for each task.
• Experimentation and Performance Testing: Lead the design and execution of experiments to rigorously evaluate models. Apply cross-validation, hyperparameter tuning, and model selection to ensure optimal performance and accuracy.
• MLOps & Workflow Optimization: Apply best practices in MLOps, including data versioning, experiment tracking, and model registries, to ensure reproducible, and efficient machine learning workflows.
• Collaborative Integration: Work closely with multidisciplinary teams, including software engineers and signal processing experts, to integrate machine learning models seamlessly into production systems.
• Technology Watch: Stay on top of the latest trends and advancements in machine learning, signal processing, and data science. Proactively identify innovative techniques that can advance our ultrasound cameras.