Staff ML Engineer, Manipulation

Hexagon Robotics, a division of Hexagon, is a global leader in precision measurement. The company specializes in developing humanoid robots for industrial sectors, aiming to address labor shortages and transition from automation to autonomy. Their first humanoid, AEON, was launched in June 2025 and is already in pilot programs with five customers. As a Staff ML Engineer, you will be part of a rapidly growing robotics team in Zurich. The team is dedicated to creating a next-generation robotic platform that automates industrial tasks with unmatched precision and intelligence. Your main responsibilities will involve developing algorithms that leverage generative AI models for intelligent object manipulation. This includes creating algorithms, training methods, and data pipelines to enable new robot skills. You will work closely with roboticists, controls engineers, computer vision engineers, and software engineers to integrate machine learning models into robust production systems. These systems are designed to automate complex industrial tasks, ensuring that the robots can interact with the physical world and perform valuable tasks for customers. To excel in this role, you must have a strong educational background and relevant experience. A Master's or PhD in Computer Science, Mechanical Engineering, Robotics, or a related field is required. Additionally, you need at least six years of experience in developing and deploying multi-modal machine learning models, preferably within the robotics domain. Proficiency in machine learning frameworks such as PyTorch, TensorFlow, or JAX, as well as solid programming skills in Python, are essential. Knowledge of robotics frameworks and tools, particularly ROS2 and Omniverse, is also beneficial. Nice-to-have skills include expertise in developing and training generative models for robot control using transformers, diffusion, or flow matching. Experience deploying machine learning algorithms on resource-constrained hardware for online inference is also advantageous. Basic German skills are a plus, but proficiency in English is required. The role offers a dynamic and collaborative environment, with a multicultural and interdisciplinary team. The company values team-oriented problem-solving and creative thinking to push the boundaries of cutting-edge technologies. The location in Zurich provides a vibrant and international culture, with a flat hierarchy where your voice counts. The company offers flexible working hours, a hybrid work model, and generous vacation time. Additional benefits include a mobility credit for sustainable commuting, a bonus system, strong pension contributions, and tailored training and development opportunities. Relocation support is also available to ensure a smooth start. Team events and discounts on health, mobility, and entertainment further enhance the work experience. What the role asks for: - Master's or PhD in Computer Science, Mechanical Engineering, Robotics, or equivalent - 6+ years of experience in developing and deploying multi-modal machine learning models - Proficiency in machine learning frameworks such as PyTorch, TensorFlow, or JAX - Solid programming skills in Python - Proficiency in English - Knowledge of robotics frameworks and tools (particularly ROS2, Omniverse) (nice to have) - Expertise in developing and training generative models for robot control (nice to have) - Experience deploying machine learning algorithms on resource-constrained hardware (nice to have) - Basic German skills (nice to have)