About the job
Meta is seeking Research Engineers to join the Multimodal Embodiment Trust team within Meta Superintelligence Labs, dedicated to advancing the safe development and deployment of Superintelligent AI. Product & Applied Research group is focused on building AI-powered experiences for people, bringing frontier models to consumers. Our two primary goals are: to build a superintelligent personal sidekick that billions of people use to make their lives better; and to provide fresh, personal, insightful entertainment by allowing people to make, share, and consume AI-generated media and immersive experiences.
Responsibilities
Design, implement, and evaluate novel, systemic, and foundational safety techniques for large language models and multimodal AI systems
Create, curate, and analyze high-quality datasets for safety system and foundations
Fine-tune and evaluate LLMs to adhere to Meta’s safety policies and evolving global standards
Contribute to applied research through risk analysis, experimentation, measurement, and and building mitigations
Work closely with researchers, engineers, and cross-functional partners to integrate safety solutions into Meta’s products and services
Qualifications
Minimum
Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
PhD in Computer Science, Machine Learning, or a relevant technical field
Experience in LLM/NLP, computer vision, or related AI/ML model training
End-to-end experience working on complex technical projects
Publications at peer-reviewed conferences (e.g. ICLR, NeurIPS, ICML, KDD, CVPR, ICCV, ACL)
Programming experience in Python and hands-on experience with frameworks such as PyTorch
Preferred
Hands-on experience applying state-of-the-art techniques to build robust AI system solutions for safety and policy adherence
Experience developing, fine-tuning, or evaluating LLMs across multiple languages and modalities (text, image, voice, video, reasoning, etc)
Demonstrated experience to innovate in safety foundational research, including custom guideline enforcement, dynamic policy adaptation, and rapid hotfixing of model vulnerabilities
Experience designing, curating, and evaluating safety datasets, including adversarial and borderline prompt cases
Experience with distributed training of LLMs (hundreds/thousands of GPUs), scalable safety mitigations, and automation of safety tooling