onsite
Sr. Research Scientist, Scaling Team
Sr. Research Scientist, Scaling Team
As a Sr. Research Scientist on the Scaling team, you will advance the scientific frontier in deep learning by developing new techniques and algorithmic innovations for foundation model efficiency, large-scale neural network training, and ML system optimization. You will lead independent research agendas and implement solutions to empower customers in applying state-of-the-art LLMs and AI systems.
About the role
About the Role
As a Sr. Research Scientist on the Scaling team at Databricks AI Research, you will be at the forefront of advancing the scientific frontier in deep learning. You will create new techniques that go beyond the state of the art, working collaboratively with a diverse team of researchers and engineers. Your primary goal will be to empower customers to successfully apply state-of-the-art LLMs and AI systems by encoding your scientific expertise into our products.
The Impact you will have
As a Sr. Research Scientist on the AI Research Team at Databricks, you will:
- Define and lead independent research agendas on foundation model efficiency in model training and reinforcement learning, conducting experiments to empirically validate hypotheses and benchmark against state-of-the-art approaches.
- Drive algorithmic innovations for large-scale neural network training or inference (e.g., novel optimizers, low-precision techniques, model adaptation methods).
- Optimize ML systems for distributed training, memory efficiency, and compute efficiency through hands-on implementation.
What We Look for
- MS/PhD in Computer Science or related field with strong foundations in machine learning and systems.
- Proven ability to write high-quality, efficient code in Python and PyTorch for research implementation and experimentation.
- Strong preference for candidates with first-author publications at top ML/systems conferences (ICLR, ICML, NeurIPS, MLSys) focused on optimization or efficiency.
Skills
Deep LearningPythonPyTorchMachine LearningDistributed Trainingmemory efficiencycompute efficiencyOptimizationneural network trainingReinforcement Learning