Research Engineer - Pretraining

Highlights

• Scale next open models from experiments to full runs at 1-3T parameters.
• Build open models for long-horizon tasks like autonomous research.
• Own pretraining runs end to end, from architecture to resilient execution.

What they offer

• Equity incentives
• Flexible work arrangements with remote or in-person option
• Visa sponsorship and relocation assistance
• Quarterly team off-sites, hackathons, conferences and learning opportunities

Ideal candidate

Hands-on model trainer who thrives owning long runs end to end and digging into training dynamics. Pragmatic, curious researcher excited to push open frontier models and collaborate across systems and data.

Requirements

• Hands-on experience training language models, including modifying training code and following effects through actual runs
• Strong Python skills and experience with PyTorch or comparable framework with systems understanding to debug distributed training
• Practical understanding of model architectures and optimization, including effect on convergence
• Ability to design an ablation that answers a research question and explain what results do and don't show
• Good judgment about what to investigate independently and when to bring in a colleague