Research Engineer - Pretraining (Fully Remote)

Prime Intellect, Inc.; United States; Hybride; USD 150 000 - 350 000; Plein temps

Avantages offerts par ce poste

Remote work

In-person in SF

Visa sponsorship

Relocation assistance

Quarterly off-sites

Learning opportunities

Résumé du poste

Prime Intellect is seeking a researcher to take open models from architecture experiments through full pretraining runs in a flexible work setup. You will work across the model and training stack, making research decisions and fixing systems problems.

Responsibilities include building and extending pretraining code, evaluating architecture changes, ensuring checkpointing resilience, and running controlled ablations with data and evaluation researchers.

Qualifications

• Hands-on experience training language models and modifying training code.
• Strong Python skills and experience with PyTorch.
• Practical understanding of model architectures and optimization.

Responsabilités

• Build and extend pretraining code, evaluate architecture changes, and profile training jobs.
• Make long runs resilient with reliable checkpointing and recovery.
• Run controlled ablations and work with data and evaluation researchers.

Connaissances

Python

PyTorch

Model architectures

Research methods

Description du poste

• Take Prime Intellect’s next open models from architecture experiments through full pretraining runs.
• Work across the model and training stack, making research decisions and fixing systems problems.

Responsibilities:

• Build and extend pretraining code, evaluate architecture changes, and profile training jobs.
• Make long runs resilient with reliable checkpointing and recovery.
• Run controlled ablations and work with data and evaluation researchers.

Requirements:

• Hands-on experience training language models and modifying training code.
• Strong Python skills and experience with PyTorch.
• Practical understanding of model architectures and optimization.

Benefits and Perks:

• Cash compensation range of $150-350k, including equity incentives.
• Flexible work arrangements, with the option to work remotely or in-person in San Francisco.
• Visa sponsorship, relocation assistance, quarterly off-sites, and learning opportunities.

Prime Intellect

Prime Intellect is building an open superintelligence stack, providing the infrastructure and platform for frontier AI training and deployment. The company has raised $150M from top investors and boasts a mission-driven team working to decentralize AI.