Senior Reinforcement Learning Engineer (Robotics)

Location: Hybrid work in Zurich, Austin, or Oxford

Company

hirify.global is a Series A robotics start-up developing Physical AI that turns heavy construction machines into autonomous robots.

What you will do

• Develop data-driven planning and control systems for autonomous excavation that generalize across machine models and soil conditions.
• Improve simulation environments and reduce the sim-to-real gap.
• Define data collection and curation pipelines for training reinforcement learning policies with real-world data.
• Design experiments to improve system performance and robustness, including adaptive and online reinforcement learning for deployed systems.
• Integrate learned components into the broader software stack and build tools to analyze their behavior.
• Collaborate with excavation and motion-planning engineers while mentoring junior team members, interns, and students.

Requirements

• 2–5 years of experience developing reinforcement learning systems for control or planning and deploying them on real robots for customers.
• Experience with GPU-accelerated simulation environments such as IsaacSim, IsaacLab, CARLA, or MuJoCo.
• Strong Python skills and experience with PyTorch or similar libraries.
• Proficiency in C++ and ability to debug real-world system behavior.
• Willingness to travel as required by business projects.

Nice to have

• Experience with hydraulic machinery, supervised learning, imitation learning, or reinforcement learning research.
• Experience deploying robotic systems at scale and working with ROS or similar robotics frameworks.
• Experience with feature-flagged deployments, staged rollouts, long-lived platforms, or ML data curation.
• Experience mentoring junior colleagues, students, or project teams.
• Familiarity with or interest in AI coding tools.

Culture & Benefits

• Work on autonomous excavation systems intended for long-term deployment across the construction industry.
• Collaborate in a growing team transitioning from prototyping to product development.
• Operate in a culture focused on thoughtful design, creative thinking, mutual respect, and pragmatism.
• Address imperfect data, noisy measurements, and the differences between simulated and physical environments.