Applied RL Researcher — Open Frontier AI (Remote)

Prime Intellect, Inc.; New York (NY); Hybrid; USD 150.000 - 300.000; Vollzeit

Benefits dieser Stelle

Professional Development budget

Equity incentives

Flexible work arrangements

Zusammenfassung

Prime Intellect, Inc. is building the open frontier AI stack for researchers and enterprise teams.

The role focuses on owning applied RL programs from customer problems through a deployed improvement loop, interfacing with customer researchers, engineers and leaders, and delivering measurable impact. You will design environments, evaluations, reward signals, and post-training experiments, and determine generalization into Prime's shared stack, with responsibilities across technology, science,

Qualifikationen

• Experience turning a workflow into tasks, environments, evaluations, and reward signals.
• Able to run post-training experiments and deploy results with measurable impact.
• Strong Python skills and ability to build artifacts that others can run and extend.

Aufgaben

• Applied Research at Prime: turn a workflow into tasks, environments, evaluations, and reward signals.
• Run post-training experiments against baselines and deploy what works.
• Determine whether gains come from model, data, harness, inference, or infrastructure.
• Make reusable parts into shared research, open-source systems, and product capabilities.
• Externally facing role with customers, users, or research partners while keeping a research agenda.
• Own deployment and adoption decisions; carry artifacts back to Prime's teams.

Kenntnisse

Python

Experimental judgment

External collaboration

Post-training ML

Build artifacts

Jobbeschreibung

Prime Intellect, Inc. is building the open frontier AI stack for researchers and enterprise teams.

The role focuses on owning applied RL programs from customer problems through a deployed improvement loop, interfacing with customer researchers, engineers and leaders, and delivering measurable impact. You will design environments, evaluations, reward signals, and post-training experiments, and determine generalization into Prime's shared stack, with responsibilities across technology, science,