Prosaic Intelligence
All roles

Research · San Francisco · In person · Internship

Member of Technical Staff, Intern

Work alongside a team of elite researchers and engineers on benchmarks, evaluations, and everything that goes into making AI safer for consumers.

About the role

Interns do the same work as the rest of the team. You will own a real project, from the question to the published result, with close mentorship, in one of the areas below.

What you might work on

Evals
Benchmarks for consumer harms, the rubrics behind them, and the LLM judges that grade them, validated against human labels.
Infrastructure
Orchestration across frontier model providers, simulated users and multi-turn environments, and the systems that take every evaluation from transcript to a reproducible, published score.
Forward deployed
Some of the team work embedded with the labs and companies we evaluate for, in tandem with their internal teams, building evaluations around their products and policies and turning what we find into fixes.

What you’ll do

  • Own a research project end to end, scoped with the team.
  • Build and run evaluations across frontier models.
  • Write up what you find, and present it to the team.

What you’ll bring

  • Research already under your belt. Authorship on a paper at NeurIPS, ICML, ICLR, ACL or a comparable venue.
  • A record that stands out. You are studying, or recently graduated, in computer science, mathematics or a related field, with a record that shows it in research, awards, or what you have built.
  • Evals you have built. You have built an evaluation, benchmark or dataset, in research or in production.
  • Strong engineering. Fluent Python, and projects that prove it, such as research code others use, open source with real users, or an internship on a serious engineering team.
  • Rigor. You don’t trust a number until you have tried to break it, and you can draw clear conclusions from messy results.
  • Ownership and speed. You can take a project from a question to a result in a few months, and you move fast without waiting to be told what to do next.

Nice to have

  • An AI safety research fellowship, such as MATS or the Anthropic Fellows Program.
  • Exceptional distinctions, such as olympiad medals or top competitive programming results.
  • A previous research or engineering internship at an AI lab.