Research Lead, Open Benchmarks Grants
Snorkel AINew York City, NY (Hybrid); San Francisco, CA (Hybrid); United States (Remote)Remote
Needs a free Superjobs account. You'll finish on Snorkel AI's Greenhouse page.
Skill score
Summary
This Research Lead shapes which AI evaluation research gets funded under Snorkel's Open Benchmarks Grants, reviewing proposals and helping grantees and Research Fellows define and strengthen benchmarks. They also work directly with frontier lab researchers to understand their measurement needs and integrate the benchmarks. The role suits a senior evaluation researcher with strong experimental rigor, hands-on implementation skills, and the ability to lead through collaboration.
- Estimated salary
- Est. $150k–224k/yr
- Estimated for a manager role, ~7 yrs experience (typical for the level), United States. Not published by the employer.
- Work mode and location
- Remote, New York City, NY (Hybrid); San Francisco, CA (Hybrid); United States (Remote)
Job description
Overview
About Open Benchmarks Grants
Open Benchmarks Grants is Snorkel’s $30M commitment to fund and support researchers and independent builders developing new AI evaluations. We support the teams behind Terminal-Bench, OSWorld 2.0, and Senior SWE-bench, and run a Research Fellows program for PhD students and postdocs. The program funds work in areas including physical AI, scientific discovery, cybersecurity, safety, and alignment, and helps teams test and strengthen their benchmarks before public release.
Roles and responsibilities
- Shape our research bets in open benchmarks & evaluation. Identify gaps in how frontier AI is evaluated. Review proposals with Snorkel’s senior researchers, work with prospective collaborators to strengthen their ideas, and help decide where we allocate our Open Benchmark Grants.
- Advance frontier evaluation research with our research partners. Work alongside grantees, Research Fellows, and Snorkel research leads to define what each benchmark measures, design tasks and experiments, and turn learnings into artifacts that other researchers can build on.
- Drive adoption and scientific exchange with frontier labs. Work directly with lab researchers to understand what they need to measure, help them integrate and interpret the benchmarks we support, and use their findings to inform subsequent research.
Skills required
Top skills
- AI evaluation
- benchmark design
- experimental design
- research collaboration
All skills (4)
- technical communication
- model behavior analysis
- verifier debugging
- evaluation harnesses
Qualifications
- A research track record in evaluation. You’ve led or made substantial contributions to benchmarks, evaluation methods, or environments that researchers or model-development teams use.
- Technical judgment about quality & experimental rigor. You can assess whether an evaluation measures what it claims, recognize alternative explanations for a result, and design experiments that resolve the uncertainty. You know which limitations matter enough to change the research direction.
- Hands-on technical depth. You’re comfortable moving between research questions and implementation: building experiments, inspecting tasks and harnesses, debugging verifiers, and analyzing model behavior.
- The ability to lead through research collaboration. You’ve helped other researchers make difficult technical decisions and bring work to completion.
- Clear technical communication. Your papers, research artifacts, specifications, or reviews make complex ideas understandable and claims precise.
Other details
Actual compensation will be determined based on factors including skills, qualifications, experience, and geographic location.
Salary range(s) for this role
$231,000—$390,000 USD
Company info
About Snorkel AI
Snorkel AI is the frontier AI data lab, helping teams build the data and environments behind high-performing frontier and agentic AI. We combine technology with research-driven AI data development to create datasets, benchmarks, evals, and custom solutions for real-world AI systems. Founded out of the Stanford AI Lab in 2019,