Tech Stack

Description

You will work end-to-end across the full ML stack, identifying and addressing key infrastructure blockers on the path to scientific AGI. You'll design and implement large-scale systems for AI scientist training, evaluation, and deployment, collaborating with researchers to scale experimental ideas.

Requirements

• 6+ years in infrastructure engineering with large-scale distributed systems expertise
• Deep knowledge of performance optimization and high-throughput ML workloads
• Experience with Docker, Kubernetes and orchestration at scale
• Proven track record of building large-scale data pipelines and distributed storage systems
• Experience with language model training infrastructure and distributed ML frameworks (PyTorch, JAX, etc.)

Responsibilities

• Design and implement large-scale infrastructure systems for AI scientist training, evaluation, and deployment
• Identify and resolve infrastructure bottlenecks impeding progress toward scientific capabilities
• Develop robust evaluation frameworks for measuring progress towards scientific AGI
• Build scalable VM/sandboxing/container architectures for safe AI task execution
• Develop large scale data pipelines for language model training