Research Engineer - Meta Superintelligence Labs (Technical Leadership)

Meta•Menlo Park, CA

1d•$219,000 - $301,000

About The Position

Meta is seeking a hands-on technical leader to advance Personal Superintelligence within Meta Superintelligence Labs (MSL). This role focuses on turning ambiguous, real-world product problems into shipped systems and measurable improvements in frontier-model capabilities, including long-horizon agents, tool use, full-stack coding, search, and personalization. In this role, you will work across the full research-to-product stack—from user experiences, backend systems, and agent environments to data, evaluations, post-training, and deployment, collaborating across research, product, design, engineering, data, and infrastructure to identify high-leverage opportunities and deliver tangible impact within clear timelines.

Requirements

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
Bachelor’s or Master’s degree in a relevant technical field, or equivalent practical experience
6 years of experience in machine learning engineering, AI research, software engineering, or a related field
4 years of providing technical leadership for complex, multi-person projects
Experience in developing or improving frontier-quality large language models or related foundation models
Deep, hands-on experience with state-of-the-art LLM post-training, data generation, evaluation, experimentation, and model behavior analysis
Track record of solving ambiguous, real-world problems and delivering measurable impact within defined timelines
Ability to work independently, lead across functions, and adapt quickly as evidence and priorities evolve

Nice To Haves

Publications at leading peer-reviewed venues such as NeurIPS, ICML, ICLR, ACL, or EMNLP, or equivalent have demonstrated industry impact in AI
Experience taking model capabilities from research prototypes to production products
Experience with large-scale distributed systems and high-throughput data pipelines
Experience developing agent harnesses, realistic environments such as web browsers or coding sandboxes, and associated evaluations
Extensive experience with long-horizon agents, agentive coding, tool use, personalization, or search

Responsibilities

Work hands-on across the full LLM post-training stack
Build high-quality training data and design and run rigorous, product-relevant evaluations
Execute, analyze, and iterate on large-scale post-training runs
Own end-to-end model capability hill-climbing, from identifying gaps through data, training strategy, evaluation, deployment, and product feedback
Advanced long-horizon agent capabilities, including tool use, full-stack coding, search, planning, and personalization
Develop realistic harnesses, environments, and evaluations for agentive tasks spanning code understanding, implementation, testing, debugging, and tool use
Research improved training, evaluation, synthetic-data generation, and data curation strategies
Translate ambiguous user and product needs into tractable research questions, technical plans, and measurable outcomes
Lead complex cross-functional projects end-to-end while remaining deeply involved in implementation, experimentation, and analysis