We are seeking a Research Scientist who brings both scientific curiosity and engineering rigor. You’ll work at the heart of the ML research pipeline—training advanced models and shaping the underlying training codebase and infrastructure.
This role suits someone passionate about reliable, scalable experimentation, large-model training, and tight integration with MLOps and systems engineering.

REQUIREMENTS

• Experience in training ML models at scale using frameworks like PyTorch or JAX
• Proficiency in Python and familiarity with deep learning optimization internals
• Understanding of GPU compute fundamentals and training-performance bottlenecks
• Strong track record in working on real datasets with practical ML applications
• Commitment to clean, maintainable, and reproducible research code

• Develop and maintain high-performance training pipelines for deep learning research
• Tune and debug large model training runs across clusters and GPU environments
• Work closely with systems engineers to ensure infrastructure supports rapid experimentation
• Collaborate with other research scientists to evolve architecture and optimization techniques
• Help establish best practices for model reproducibility, logging, and deployment readiness