About the role#
As a Research Fellow, you will build open benchmarks and evaluation environments that shape how AI progress is measured. This position is for PhD students and postdocs who want to turn important questions in their fields into new ways to evaluate AI capabilities. You will own the project from the initial research question through public release. You will work with a Snorkel researcher as a co-advisor and have access to resources including domain experts and compute for your experiments. You will retain your academic affiliation while spending at least one day a week with the research team.
What you'll do#
- Scope and lead a new benchmark, dataset, or environment in a domain where current evaluations fall short.
- Define task specifications, rubrics, and reference answers alongside domain experts.
- Run the expert review process.
- Release the research artifact publicly as the lead author.
- Join weekly research meetings to share progress, receive feedback, and contribute to other projects.
What you'll need#
- You are a PhD student or postdoc with a research background in designing and building open benchmarks, datasets, or environments.
- Demonstrated ability to turn research ideas into working code, datasets, or tools.
- Ability to work with domain experts outside your field and turn their judgment into a specification.
- Ability to scope a project to a public release within the fellowship term.
- Experience in Computer Science.
Preferred qualifications include:
- Publications at venues such as NeurIPS, ICML, ICLR, or top venues in your discipline.
- Experience with agentic evaluation, RL environments, or LLM evaluation harnesses.
- Deep expertise in a domain where expert-level evaluation is scarce, such as the sciences, medicine, law, finance, or engineering.
Location & details#
- This is a full-time, paid position.
- The role is available as a hybrid position in New York City or San Francisco, or as a remote role within the United States.
- The term is rolling.
About Snorkel AI
Snorkel AI develops software for the data layer of artificial intelligence systems. The company serves government agencies, enterprises, and research labs. Its technology helps teams move machine learning projects from initial prototypes into production environments. Founded in 2019, the firm is based in Redwood City, California. It employs between 51 and 200 people.
How to get in at Snorkel AI
Applying early to Snorkel AI gives you a better chance of having your resume reviewed before the applicant pool grows too large. Intern Insider sends an instant alert the moment a role matching your target is published anywhere, so you can be among the first to submit your materials. This timing is a real advantage in a competitive market. You can also improve your response rates by reaching out to the recruiters behind these roles directly. Intern Insider surfaces the specific people managing hiring for Snorkel AI, allowing you to ask relevant questions or request a referral. Connecting with a human is often more effective than waiting for a response from an automated system.


