Research Scientist: Multi-Modal AI & World Models

Meta; Pittsburgh, Redmond (Allegheny County, WA); On-site; USD 180,000 - 300,000; Full time

Job summary

Meta is seeking a Research Scientist to advance foundational research in multi-modal understanding, synthesis, and world models. The role focuses on building AI systems that perceive, reason, and generate content across vision, language, audio, and other modalities.

You will lead research initiatives, publish in top venues, and collaborate with engineers to translate breakthroughs into Meta's next-generation AI products.

Qualifications

• PhD in ML, CV, NLP or related field, or equivalent practical experience.
• 6+ years conducting AI research in multi-modal learning or world models.
• Experience leading research initiatives from conception to publication or deployment.
• Proficiency in Python and deep learning frameworks (PyTorch/TensorFlow).
• Strong publication record in top ML/AI venues.

Responsibilities

• Lead foundational research in multi-modal understanding, synthesis, and world models.
• Develop and evaluate large-scale multi-modal systems and models.
• Publish influential work and present findings to diverse audiences.
• Collaborate with research and engineering teams to deploy breakthroughs.

Skills

Multi-modal learning

World models

Deep learning

Python

PyTorch

TensorFlow

Research leadership

Publications

Cross-functional comms

Model-based RL

Education

PhD in ML / CV / NLP

Bachelor's in CS/EE or equivalent

Tools

PyTorch

TensorFlow

Job description

Meta is seeking a Research Scientist to advance foundational research in multi-modal understanding, synthesis, and world models. The role focuses on building AI systems that perceive, reason, and generate content across vision, language, audio, and other modalities.

You will lead research initiatives, publish in top venues, and collaborate with engineers to translate breakthroughs into Meta's next-generation AI products.