Research Scientist Intern (AI Infrastructure)

TL;DR

Research Scientist Intern (AI Infrastructure): Designing and optimizing scalable AI infrastructure and distributed systems with an accent on ETL pipelines, GPU/TPU scheduling, and orchestration. Focus on building fault-tolerant deployment systems, profiling the ML stack, and translating research prototypes into production-ready infrastructure.

Location: Seattle, Washington, United States

Salary: USD 156,000 – 316,000 (estimate)

What you will do

• Architect ETL and data ingestion pipelines and design scalable AI infrastructure.
• Build distributed deployment, orchestration systems, and implement containerized architectures.
• Implement workflow orchestration to transition projects from research to production.
• Develop telemetry and benchmarking frameworks to profile and optimize the ML stack.
• Improve fault tolerance and high availability while integrating experiment management tools.
• Mentor engineers on reliability engineering and performance tuning.

Requirements

• PhD candidate or graduate.
• Experience with distributed systems, Kubernetes, and Docker.
• Knowledge of ML compilers, GPU scheduling, and TPU scheduling.
• Proficiency in data engineering and orchestration tools (e.g., Apache Spark, Airflow).
• Ability to translate prototype requirements into production systems.