Tech Stack

Description

You will develop advanced statistical models and machine learning methodologies to optimize GPU utilization and infrastructure efficiency. You'll analyze large-scale system telemetry data and prototype predictive and optimization algorithms that directly inform production systems, turning complex infrastructure data into measurable improvements in performance and cost.

Requirements

• MS or PhD in Computer Science, Statistics, Applied Mathematics, Machine Learning or related field
• 8+ years applying statistical modeling or ML to large-scale datasets
• Strong proficiency in Python and scientific computing libraries (NumPy, pandas, scikit-learn, PyTorch or TensorFlow)
• Experience with distributed data systems (Spark, Ray, Dask) and SQL
• Experience designing and analyzing controlled experiments (A/B testing, causal inference)

Responsibilities

• Develop statistical models and ML methodologies to optimize GPU utilization and infrastructure efficiency
• Design experiments and analyze large-scale system telemetry data
• Prototype predictive and optimization algorithms for production systems
• Collaborate cross-functionally to translate research insights into deployable solutions
• Improve cluster reliability, increase effective utilization, and reduce MTTR