Close the capability overhang between power users and average consumers — build evals, training data, and reward signals that make ChatGPT more capable for everyone.

OpenAI's Personal AGI North Stars team is hiring a Research Engineer/Scientist to improve model capability and behavior across tool-use, feature discovery, connectors, and instruction following, deployed to millions of ChatGPT/API users. You'll own a research agenda, build evals to track capability improvements, and work across a large ML codebase to translate behavioral bottlenecks into training data and reward signal changes. This interview probes ML research judgment, eval design rigor, and hands-on ability to debug and iterate in a complex research stack.

Practice this interview

Free · a live voice mock calibrated to this exact role

Start the mock interview

What this interview tests

Identifying and closing model capability bottlenecks (tool-use, instruction following, feature discovery)
Building robust evals that track real capability/behavior improvements
Translating eval findings into training data and reward signal changes
Debugging and iterating within a large, complex ML research codebase
Balancing independent research ownership with cross-team collaboration

Common question themes

Walk through how you'd design an eval for a specific model capability gap

Describe translating an eval result into a concrete training data or reward signal change

Tell me about debugging a subtle behavior issue in a large ML codebase you didn't build

How do you avoid an eval being gamed or failing to generalize

How would you prioritize a research agenda that touches tool-use, connectors, and instruction following simultaneously

How candidates describe it

Real Research Engineer/Research Scientist interview stories — retold from candidates' public write-ups, with sources.

Amazon · Applied Scientist (L4)Offer

Amazon Applied Scientist Interview Experience: Alexa Speech Team, 2021

A redirected recruiter call turned into an Amazon Applied Scientist loop with the Alexa Speech team: a phone screen, a split five-round virtual onsite across two teams, a bar raiser, and an added ML-breadth round, ending in an offer with a downlevel from L5 to L4.

Interviewed July 2021 · Boston, MA (Remote)

Google · L5 Machine Learning EngineerOffer

Google L5 machine learning engineer interview: phone screen skipped, four technical rounds, and an offer

A candidate applying for an L5 machine learning engineer role at Google had the phone screen skipped due to a referral and prior tenure at the company, then went through four technical and design rounds plus a behavioral round before receiving an L5 offer. The loop was one leg of a broader search that produced offers from several companies in the same cycle.

Interviewed April 2022 · Remote

View the original posting

All OpenAI interviews