Join the team.

We're building physics-informed foundation models — and we need someone who can own the full arc from research to production. You'll work on our core time-series and vision models, design training infrastructure, and make sure what works in a notebook works in the real world.

You Will

• Develop and iterate on core time-series and vision models
• Design training and inference infrastructure (GPU orchestration, experiment tracking, model serving)
• Bridge research to production — turn prototype code into deployable systems
• Drive MLOps practices: versioning, monitoring, CI/CD pipelines

You Bring

• Deep experience with time-series or vision models in an R&D setting
• Track record shipping research code as production services
• Proficiency with PyTorch and experiment tracking (W&B, MLflow)
• Experience managing cloud GPU training pipelines (GCP, AWS, or Azure)

Bonus

• Distributed training frameworks
• ML deployment in regulated or industrial environments