SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
OpenAI's Recursive Self-Improvement (RSI) team is hiring research engineers, research scientists, and AI systems engineers to work on automating research workflows and accelerating research productivity through AI systems.
In this role, you will design evaluations for research judgment, hypothesis generation, and long-horizon experiment execution. You'll turn real research workflows and model failures into data and evaluation flywheels, improving model research capabilities through agent harnesses, synthetic data, RL environments, and model training. You'll build and maintain safe, reliable integrations between OpenAI's models and research infrastructure, develop research agents and experiment-orchestration systems, and create metrics and economic models to understand the impact on research productivity and model capabilities.
This is a high-ownership role for researchers and engineers who thrive in ambiguity, move fluidly between research and implementation, and turn emerging opportunities into rigorous, reliable, scalable results.
You should have research or engineering experience across LLM training, model evaluations, agent systems, synthetic data, research infrastructure, or large-scale distributed systems. You're a strong generalist who can move between open-ended research and practical implementation, turning ambiguous problems into clear results. You collaborate effectively across the full stack—systems, data, model training, evaluations, and other research teams—and are comfortable building and maintaining data pipelines, tooling, and infrastructure needed to support emerging AI capabilities. You think rigorously about scientific quality, research taste, safety, privacy, reliability, performance, and scale, and are excited about using increasingly capable AI systems to accelerate meaningful research.
