Open-Frontier AI Research Engineer (Remote)

Prime Intellect, Inc.; San Francisco (CA); Remoto; USD 150.000 - 350.000; Tempo pieno

Vantaggi offerti da questo lavoro

Cash compensation

Remote or SF office

Visa sponsorship

Team off-sites

Learning opportunities

Mission-driven team

Descrizione del lavoro

Prime Intellect, Inc. is building an open frontier AI platform that unifies compute, environments, evaluations, and secure sandboxes for end-to-end frontier-scale training and deployment.

The role focuses on taking open models from architecture experiments to full pretraining runs, coordinating across model and training stacks, and solving system-level challenges to ensure successful runs. You will work across the training stack, evaluating architecture changes, profiling jobs, and making runs

Competenze

• Hands-on experience training language models and modifying training code.
• Strong Python skills and debugging distributed training systems.
• Practical understanding of model architectures and optimization.
• Ability to design ablations to answer research questions.

Mansioni

• Build and extend the pretraining code in prime-rl, taking small experiments into full training runs.
• Evaluate architecture changes (MoE, attention) for quality and cost.
• Profile training jobs and improve throughput with hardware-aware execution.
• Ensure long runs are resilient with reliable checkpoints and recovery.
• Investigate training dynamics and stability to preserve convergence.
• Run controlled ablations and keep setups reproducible.
• Collaborate with data and evaluation researchers on mixtures and long-context experiments.

Conoscenze

Language models

Python

PyTorch

Distributed training

Model architectures

Ablations

Experiment design

Strumenti

PyTorch tooling

Distributed training infra

Descrizione del lavoro

Prime Intellect, Inc. is building an open frontier AI platform that unifies compute, environments, evaluations, and secure sandboxes for end-to-end frontier-scale training and deployment.

The role focuses on taking open models from architecture experiments to full pretraining runs, coordinating across model and training stacks, and solving system-level challenges to ensure successful runs. You will work across the training stack, evaluating architecture changes, profiling jobs, and making runs