Research Engineer, Discovery

About the role

Anthropic is seeking a senior Research Engineer to join its Discovery team in San Francisco. You will work across the full model stack, tackling infrastructure challenges that stand between current capabilities and scientific AGI. The role requires rapid learning and hands‑on implementation in areas such as performance optimization, distributed systems, and large‑scale data pipelines.

Key responsibilities

• Design and implement large‑scale infrastructure for AI scientist training, evaluation, and deployment.
• Identify and resolve bottlenecks that hinder progress toward scientific capabilities.
• Develop robust evaluation frameworks to measure advances toward scientific AGI.
• Build scalable VM, sandbox, and container architectures for safe execution of long‑horizon AI tasks.
• Translate experimental requirements into production‑ready infrastructure.
• Create and maintain large‑scale data pipelines for advanced language model training.
• Optimize training and inference pipelines for stable, efficient reinforcement learning.

Required profile

• 6+ years of experience in infrastructure engineering with large‑scale distributed systems.
• Strong communicator who thrives in collaborative, fast‑paced environments.
• Proven ability to diagnose and resolve complex production infrastructure issues.
• Experience working across the full ML stack, from data pipelines to performance tuning.
• Track record of collaborating with researchers to scale experimental ideas.

Required skills

• Docker
• Kubernetes
• Containerization technologies
• Performance optimization techniques for high‑throughput ML workloads
• Design of large‑scale distributed systems
• Building and operating large‑scale data pipelines
• Distributed storage systems
• Reinforcement learning pipeline optimization