Multimodal AI Research Scientist: Video & Personalization

lumalabs-ai; New York (NY); On-site; USD 200,000 - 450,000; Full time

Job summary

Luma AI in New York seeks a full-time senior applied researcher to refine and deploy video foundation models. You’ll work at the intersection of research, product, and partnerships to turn state-of-the-art techniques into practical capabilities for creative workflows.

You will drive controllability, personalization, and evaluation across modeling, data, and systems, partnering with design and product to translate user intent into robust model behavior and intuitive controls.

Qualifications

• Strong ML fundamentals with deep experience in visual generative models (diffusion/transformers or related architectures).
• Ideal candidates have expertise in fine-tuning, personalization, domain adaptation, data curation, or interpretability.
• Hands-on Python and DL engineering skills, comfortable moving between research prototypes and production systems.

Responsibilities

• Refine, personalize, and deploy video foundation models across modeling, data, systems, and evaluation.
• Develop and maintain model variants for user environments and creative partners.
• Define success metrics and build user-aligned evaluations to ensure fidelity and reliability.
• Collaborate with product, research, and design to translate creative intent into production-ready controls.

Skills

Python

PyTorch

Research

Production Systems

Education

PhD in ML/CS

Tools

Ray

Kubernetes

Slurm

Job description

Luma AI in New York seeks a full-time senior applied researcher to refine and deploy video foundation models. You’ll work at the intersection of research, product, and partnerships to turn state-of-the-art techniques into practical capabilities for creative workflows.

You will drive controllability, personalization, and evaluation across modeling, data, and systems, partnering with design and product to translate user intent into robust model behavior and intuitive controls.