Autonomous AI Researcher: RL & Training (Remote)

Prime Intellect, Inc.; Oregon; Hybrid; USD 150,000 - 300,000; Full time

Benefits offered by this job

Flexible work locations

Visa sponsorship

Relocation support

Development budget

Team off-sites & conference attendance

Job summary

Prime Intellect, Inc. in the United States is seeking researchers to build and post-train agents for scientific research, including literature search, code execution, using domain tools, and selecting experiments within fixed compute budgets.

You will work on autonomous research and computational science, with environments and evaluations built around partner tools and data, delivering reusable research code and experiments.

Qualifications

• Depth in RL, agents, or post-training systems and ability to acquire the relevant scientific context.
• Hands-on experience building and debugging model-training, post-training, agent, evaluation, or closely adjacent experimental loops.
• Strong Python and the ability to build artifacts that others can run, inspect, and extend.

Responsibilities

• Externally facing Applied Research role; often work with customers, users, or research partners.
• Build experimental loops for scientific tools and datasets; reproduce and verify results.
• Develop long-horizon environments where agents search literature, run code, query databases, and choose experiments within compute budgets.
• Turn partners' workflows and data into trainable tasks, evaluations, and reward signals.
• Design verifiers for correctness, novelty, and reproducibility; check for leaks or shortcuts.
• Connect programmatic checks, simulations, domain models, and expert feedback to training.

Skills

RL depth

Agents

Post-training systems

Python

Job description

Prime Intellect, Inc. in the United States is seeking researchers to build and post-train agents for scientific research, including literature search, code execution, using domain tools, and selecting experiments within fixed compute budgets.

You will work on autonomous research and computational science, with environments and evaluations built around partner tools and data, delivering reusable research code and experiments.