Latest AI JobsBrowse Jobs
Luma AI logo

Research Scientist - World Model

Luma AIRedwood City, CAResearch & AI
Apply on company site
Salary
Not listed
Job type
Full-time
Work mode
Remote
Posted
24 Jul 2026
Status
Open · checked 7d ago

About this role

Role Overview

Luma AI is hiring a Research Scientist - World Model to transform leading generative video models into interactive, physically faithful substrates for embodied reasoning. Located in Redwood City, California with remote flexibility, this position suits researchers with deep generative modeling or model-based reinforcement learning expertise who enjoy open-ended problem solving.

Key Responsibilities

Design and invent next-generation world-model architectures using transformer, diffusion, autoregressive, or hybrid methods to ensure physical consistency and controllability.
Create advanced controllability mechanisms including view conditioning, action conditioning, and long-horizon rollouts for agent interaction.
Establish and take ownership of critical success metrics such as action-following, physical fidelity, long-horizon coherence, and downstream policy training usefulness.
Conduct comprehensive scaling studies to determine optimal payoffs across compute, architecture, and data.
Share research findings through top-tier publications and contribute directly to open-source model releases.

Requirements

Hold a PhD or possess an equivalent research background in computer vision, machine learning, robotics, or a closely related discipline.
Bring deep specialized knowledge in self-supervised representation learning, large-scale generative modeling encompassing video, 3D, or world models, or model-based reinforcement learning.
Demonstrate strong practical skills in PyTorch alongside extensive experience training models at the limits of multi-node clusters.
Showcase a recognized research portfolio featuring widely utilized open-source contributions and/or top-venue publications.

Nice to Have

Background working directly with world models, generative video, model-based reinforcement learning, 4D scene representations, or neural simulation.
Practical experience applying generative models toward downstream embodied tasks like control, planning, and evaluation.
Passion for open-sourcing cutting-edge frontier models.

Skills and Tools

PyTorch
Computer Vision
Large-scale Generative Modeling
Self-Supervised Representation Learning
Model-Based Reinforcement Learning

About the Company

Luma AI focuses on developing unified general intelligence capable of understanding, generating, and operating within the physical world. The organization emphasizes multimodality, asserting that vision represents the essential next step beyond language models. Luma AI operates as an equal opportunity employer.

How to Apply

Interested candidates should visit the official Luma AI careers page to submit an application for this opportunity.

Skills and tech stack

Questions candidates ask

Where is the Research Scientist - World Model position located?

The role is based in Redwood City, CA, and offers remote work flexibility.

What qualifications are required for this role?

Candidates need a PhD or equivalent research record in machine learning, computer vision, robotics, or a related field, along with deep expertise in generative modeling, self-supervised learning, or model-based RL.

What technical framework experience is necessary?

Strong PyTorch skills and experience with large-scale training to the limits of a multi-node cluster are required.

What will the first 90 days look like?

The first 30 days involve immersing and diagnosing current video models, days 30 through 60 focus on prototyping and validating controllability mechanisms, and days 60 through 90 center on running scaling studies and pushing toward open release.

What is the core mission of Luma AI?

Luma AI builds unified general intelligence that can generate, understand, and operate in the physical world.

Ready to apply?

The application is on the official Luma AI careers page.

Apply on company site