GrepJob
OpenAI

Researcher: Agent Post-Training, API & Power-Users

OpenAI
Apply
about 2 hours ago
San Francisco, CA, USAMid Level / Senior
H1B Sponsor

Base Salary

$295k - $445k/yr

Responsibilities

  • Design and run experiments to improve model behavior in API and power-user workflows.
  • Build evaluations and environments based on real developer workflows.
  • Transform observed model failures into training data and hypotheses.
  • Partner with users to identify behavior gaps and implement post-training interventions.
  • Own end-to-end model behavior projects from analysis to launch readiness.
  • Develop feedback loops using user traces and API patterns to identify model gaps.
  • Decide on agentic capabilities and behavioral fixes for major model runs.
  • Debug failures in shipped models by analyzing traces and evals.
  • Work on early-training and alignment interventions to shape agent behavior.
  • Improve large-scale training processes for reliability and efficiency.

Requirements

  • Strong technical fundamentals in ML, software engineering, or applied research.
  • Hands-on experience with LLMs, post-training, and production ML systems.
  • Ability to form concrete hypotheses about model behavior from evaluations.
  • Excitement for solving ambiguous capability problems with noisy signals.
  • Deep care for developer and expert-user experience in real workflows.
  • Comfortable working across research, product, and infrastructure teams.
  • Willingness to build systems and processes as needed by the team.
  • Desire to train and ship models that enhance usability for various users.

Categories

AI & MLBackendData Science