Research Engineer, Simulation & World Models

Build simulation systems that train and evaluate robots and embodied agents. Model physical interactions, create scalable training worlds and connect simulation with real-world data across Mecka Labs. This is a hands-on research engineering role for someone exceptional in simulation, physics and learning. You'll work with MuJoCo, Isaac Sim and related tools, close sim-to-real gaps, and build learned or hybrid world models for prediction, planning and control. You own whether the simulated world behaves credibly and runs at scale; partner roles turn it into trainable tasks and reliable evaluations. **What you will be doing:** - **Build simulation environments:** Model robots, sensors, objects, contacts, materials and dynamics, with scenarios grounded in real behavior. - **Scale training workloads:** Build reliable pipelines for parallel rollouts, synthetic data, policy training and evaluation. - **Improve physical fidelity:** Calibrate against measured data, find mismatches in dynamics or sensing and make targeted improvements. - **Drive sim-to-real:** Use system identification, domain randomization and controlled experiments to improve transfer. - **Develop world models:** Build learned dynamics or latent models, combine them with physics-based systems and test their value for prediction, planning and control. - **Build with the team:** Turn prototypes into reusable simulation assets, training infrastructure and documented methods with researchers and engineers. **What you bring:** - **Simulation and physics:** Deep experience building and debugging physics-based simulation with MuJoCo, Isaac Sim or comparable platforms. - **Robotics learning:** Strong reinforcement learning and embodied AI fundamentals, including observations, actions, rewards and policy evaluation. - **World models:** Experience with learned dynamics, latent world models or hybrid physics-learning systems. - **Research engineering:** Strong Python and C++ systems skills; you build scalable, reproducible tools other researchers can extend. - **Experimental judgment:** Design controlled experiments, measure sim-to-real gaps and trace failures to models, data, policies or infrastructure. **Even better if you have:** - Simulation or training infrastructure used at scale by a robotics or embodied AI research team. - Demonstrated sim-to-real transfer in manipulation, locomotion, navigation or another physical domain. - Published or open-source work in world models, differentiable simulation, GPU-accelerated simulation or model-based reinforcement learning. **A Note on Applying** Studies show women and candidates from underrepresented groups often only apply when they meet 100% of the listed qualifications, while others apply after meeting 60%. If you don't check every box above but believe you can do the job, we encourage you to apply - we're looking for capability and trajectory, not a perfect checklist match.