About the role
The robotics of the future isn't just developed on real machines — it's built in simulation first. At NEURA Robotics, we're looking for someone who doesn't just operate complex simulation environments, but uses them systematically to objectively evaluate learning and control methods, and to pave the way from simulation to the real world.
Your mission & challenges
- Simulation, Benchmarking & Evaluation: You build, configure, and operate complex robotics simulations. You develop standardized benchmark environments and test cases, run regular in-the-loop tests, and evaluate different control, planning, and learning approaches against clearly defined metrics.
- Experiment Design & Reproducibility: You design reproducible simulation and evaluation experiments, conduct systematic ablation studies, and ensure clean versioning and documentation of scenarios, configurations, and results — including structured benchmark reports.
- Sim-to-Real & Model Validation: You analyze the transferability of simulation results to real robot platforms, identify and quantify sim-to-real gaps (e.g. dynamics, sensor noise, contact models), and calibrate simulation parameters for maximum physical realism.
- Collaboration & Interfaces: You work closely with ML, robotics, and software teams, jointly define benchmarks and quality thresholds with NEURA, and present simulation results clearly and accessibly to technical stakeholders.
What we can look forward to
- A university degree (Bachelor's or Master's) in Robotics, Computer Science, Mechanical Engineering, Mechatronics, Computational Engineering, or a related field
- At least three years of relevant professional experience in robotics simulation, benchmarking, or model-based evaluation — with proven hands-on experience in research or industry
- Strong expertise in MuJoCo (contact models, actuation, sensors, XML models) and experience with multi-GPU-accelerated simulation environments such as NVIDIA Isaac Sim or PyBullet
- Confident handling of robot and environment modelling (URDF, MJCF, USD) and a solid understanding of robotic dynamics, kinematics, and contact physics
- Experience developing objective metrics and statistically evaluating simulation results
- Knowledge of structured experiment management, Python-based toolchains, and automation of simulation tests (batch runs, sweeps, CI-like workflows)
- Experience with or strong interest in sim-to-real transfer methods, as well as a good understanding of real robot hardware and its limitations
- Team spirit, a structured working style, and the ability to document and communicate technical results with precision