Remote RL Research Engineer - Frontier AI Platform

Prime Intellect, Inc.; San Francisco (CA); Hybride; USD 150 000 - 350 000; Plein temps

Avantages offerts par ce poste

Cash compensation

Remote or in-person options

Visa sponsorship

Off-sites & conferences

Résumé du poste

Prime Intellect in San Francisco is building open frontier AI and seeking research engineers and scientists to develop RL algorithms, scaling research, and training frameworks. You will contribute to a full-stack platform unifying compute, environments, evaluations, and deployment for frontier models.

You will design controlled experiments, test new methods, and help push the boundaries of frontier AI in a collaborative research environment with opportunities to publish and impact real-world

Qualifications

• Demonstrated depth in at least one of RL algorithms, empirical research on model training, or distributed ML systems.
• Strong programming skills in Python and experience with PyTorch or a comparable ML framework.
• Ability to design controlled experiments, interpret results, and debug unexpected model or system behavior.
• Experience taking a research idea or systems change through implementation, testing, and evaluation, with a clear account of what worked and what didn’t.

Responsabilités

• Develop and test RL methods for reasoning, coding, and long-horizon tasks.
• Investigate stability in asynchronous RL, including policy staleness and trainer/sampler mismatch.
• Turn experimental findings into training recipes and test across larger models and longer runs.
• Extend the RL framework to support new algorithms and models, and improve performance with the infra team.
• Build diagnostics to reproduce unstable runs and distinguish algorithm problems from runtime issues.

Connaissances

Python

PyTorch

Experiment design

Debugging

Description du poste

Prime Intellect in San Francisco is building open frontier AI and seeking research engineers and scientists to develop RL algorithms, scaling research, and training frameworks. You will contribute to a full-stack platform unifying compute, environments, evaluations, and deployment for frontier models.

You will design controlled experiments, test new methods, and help push the boundaries of frontier AI in a collaborative research environment with opportunities to publish and impact real-world