Robotics Simulation Engineer: World Models

mecka; San Francisco (CA); In loco; USD 140.000 - 200.000; Tempo pieno

Descrizione del lavoro

Mecka AI is hiring for a hands-on research engineering role focused on simulation, physics, and learning for robotics and embodied AI. You will work with MuJoCo, Isaac Sim and related tools to build scalable, credible simulation worlds and learned or hybrid world models for prediction, planning and control.

You will own the sim quality, contribute to training pipelines, and partner with researchers to turn prototypes into reusable assets and robust evaluation platforms.

Competenze

• Deep experience building and debugging physics-based simulation with MuJoCo, Isaac Sim or comparable platforms.
• Strong reinforcement learning and embodied AI fundamentals.
• Experience with learned dynamics or hybrid physics-learning systems.
• Strong Python and C++ systems skills; scalable, reproducible tools.

Mansioni

• Build simulation environments modeling robots, sensors, objects, contacts and dynamics grounded in real behavior.
• Scale training workloads with reliable pipelines for parallel rollouts and synthetic data.
• Improve physical fidelity by calibrating against measured data and fixing mismatches.
• Drive sim-to-real transfer using system identification and domain randomization.
• Develop world models and test their value for prediction, planning and control.
• Collaborate with researchers and engineers to turn prototypes into reusable assets.

Conoscenze

Simulation

Physics

Reinforcement learning

World models

Research engineering

Python

C++

Experimental judgment

Strumenti

MuJoCo

Isaac Sim

GPU-accelerated simulation

System identification

Descrizione del lavoro

Mecka AI is hiring for a hands-on research engineering role focused on simulation, physics, and learning for robotics and embodied AI. You will work with MuJoCo, Isaac Sim and related tools to build scalable, credible simulation worlds and learned or hybrid world models for prediction, planning and control.

You will own the sim quality, contribute to training pipelines, and partner with researchers to turn prototypes into reusable assets and robust evaluation platforms.