Simulation Researcher/Engineer

lumalabs-ai; San Francisco (CA); On-site; USD 250,000 - 450,000; Full time

Job summary

lumalabs-ai is building a faithful, controllable world simulation to train general-purpose robot policies. You will operate at the boundary between generative models and physics simulation to decide where each earns its keep.

You will design visually rich, physically plausible environments, build evaluation harnesses, and develop GPU-accelerated pipelines, collaborating with world-model and policy-learning researchers to advance sim-to-real transfer.

Qualifications

• Strong background in robotics simulation, computer graphics, physics-based modeling, or generative 3D.
• Fluency in Python and C++; familiarity with a production physics engine.
• Track record of building simulation systems used by others.

Responsibilities

• Design simulation environments that are visually rich, physically plausible, and scalable.
• Build evaluation harnesses to measure world model quality for robots.
• Develop differentiable and GPU-accelerated simulation pipelines.
• Drive asset, scene, and task generation pipelines using generative stack.
• Collaborate with world-model researchers and policy-learning researchers.

Skills

Robotics simulation

Python

C++

Isaac Sim

MuJoCo

Bullet

PhysX

Drake

Tools

Isaac Sim

MuJoCo

Bullet

PhysX

Drake

Job description

THE ROLE

We're building a sufficiently faithful, controllable simulation of the world — on top of Luma's generative video and 3D models — as the substrate for training general-purpose robot policies. As a Simulation Researcher/Engineer, you'll help define what that substrate looks like. You will sit at the boundary between generative models and classical physics simulation, and you'll be one of the people who decides where each one earns its keep.

WHAT YOU'LL DO

• Design simulation environments that are visually rich, physically plausible, and trainable at scale — a hybrid of generative rollouts and physics-engine-based scenes.
• Build the evaluation harness that tells us whether our world model is good enough to train robots on (sim-to-real gap, physical consistency, long-horizon coherence).
• Develop differentiable and GPU-accelerated simulation pipelines where they unlock new training signal.
• Drive the asset, scene, and task generation pipelines — including using Luma's own generative stack to bootstrap diversity.
• Collaborate with world-model researchers (upstream) and policy-learning researchers (downstream).

MINIMUM QUALIFICATIONS

• Strong background in robotics simulation, computer graphics, physics-based modeling, or generative 3D — degree or equivalent practical record.
• Fluency in Python and C++; deep familiarity with at least one production physics engine (Isaac Sim/Lab, MuJoCo, Bullet, PhysX, Drake).
• Track record of building simulation systems other people actually used.

PREFERRED

• Research on sim-to-real transfer, domain randomization, differentiable simulation, or neural-rendering-based simulation.
• Game engine, CGI, animation, or photogrammetry background.
• Publications at top venues (CoRL, RSS, ICRA, NeurIPS, SIGGRAPH).

Compensation

The base pay range for this role is $250,000 – $450,000 per year.