AI Research Scientist, LLM Post-Training - Meta Superintelligence Labs

Meta

Location
Bellevue, WA
Pay
$184,000–$257,000 a year
Level
Mid
Posted
, on Meta's careers site

Sponsorship check

What we checked before showing you this job.

  • Meta sponsors H-1B visas

    5,998 filings in FY2026, and every year since 2020.

  • They sponsor this role

    134 Machine Learning Engineer filings in FY2026, at a median of $223,000.

  • The pay matches their H-1B filings

    This job pays $184,000–$257,000. Their H-1B machine learning engineer filings were at $201,000–$233,000.

  • E-Verify employer

    You can take this job on STEM OPT.

  • The post doesn't mention sponsorship

    Ask the recruiter early.

Always confirm with the recruiter. hire.page shows public records and what the post says, not a promise from Meta.

About the job

We are seeking AI researchers for the Modeling & ToolCalling team within Meta Superintelligence Labs. This team is focused on LLM post-training, with emphasis on advancing long-horizon agentic tool use, search, personalization, and other frontier capabilities required for models to operate effectively in real-world settings.

Responsibilities

Work across the full LLM post-training stack Building high-quality training data, and designing and running evaluations Execute and iterate on post-training runs Own end-to-end LLM capability “hill-climbing” Conduct research on improved training and data curation strategies Lead complex technical projects end-to-end

Qualifications

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience Currently has, or is in the process of obtaining, a PhD degree in Computer Science or a related technical field 2+ years of industry research experience in LLM/NLP, computer vision, or related AI/ML models Deep, practical experience in LLM post-training Experience as a technical lead on a team and/or leading complex technical projects from end to end Published research in leading peer-reviewed conferences (e.g., ACL, NeurIPS, ICML, ICLR, AAAI, KDD, CVPR, ICCV) and/or demonstrated significant industry influence in the field of AI Extensive experience working on long horizon agents, agentic tool use, personalization, and/or search Experience working on frontier-quality, state-of-the-art Large Language Models First-author publications at top peer-reviewed conferences (e.g., ACL, NeurIPS, ICML, ICLR, AAAI, KDD, CVPR, ICCV)

Job text from Meta's careers site.