Research Fellow: Build Open AI Benchmarks & Evaluations

Snorkel AI; United States; Hybrid; USD 117,000 - 227,000; Full time

Benefits offered by this job

Co-advisor matched to project

Compute resources for your project

Open-source research artifact release

Access to research community events

Job summary

Snorkel AI is seeking a Research Fellow who will lead the development of open benchmarks, datasets, or environments. You will own the project from question to public release while spending time with Snorkel’s research team.

You’ll retain your academic affiliation and work with a co-advisor, gaining access to compute, domain experts, and support for publications and open research artifacts.

Qualifications

• PhD candidates or postdocs pursuing research in AI evaluation and benchmarks.
• Ability to turn research ideas into working code, datasets, or artifacts.
• Experience collaborating with domain experts and translating judgment into specifications.

Responsibilities

• Scope and lead a new benchmark, dataset, or evaluation environment.
• Define task specifications, rubrics, and reference answers with domain experts.
• Release the research artifact publicly as lead author.
• Attend weekly research meetings and contribute to other projects.

Skills

Research expertise

Code implementation

Academic collaboration

Publications (NeurIPS/ICML)

RL environments

Agentic evaluation

Domain expertise

Education

PhD student or PhD

Job description

Snorkel AI is seeking a Research Fellow who will lead the development of open benchmarks, datasets, or environments. You will own the project from question to public release while spending time with Snorkel’s research team.

You’ll retain your academic affiliation and work with a co-advisor, gaining access to compute, domain experts, and support for publications and open research artifacts.