Staff Research Engineer - AI Discovery & AGI Enablement

Anthropic; New York (NY); Hybrid; USD 350,000 - 850,000; Full time

Benefits offered by this job

Equity donation matching

Generous vacation

Parental leave

Flexible hours

Office in SF

Job summary

Anthropic is seeking a Research Engineer to work end-to-end on bottlenecks toward scientific AGI. You will contribute to language model training, evaluation, and inference, scaling research ideas from prototype to production, while optimizing performance and deploying via VM/sandboxing and cloud infra.

Strong candidates have extensive ML research experience, familiarity with large-scale LMs, and a track record of shipping ML systems for complex multi-step reasoning.

Qualifications

• 8+ years of ML research experience.
• Familiarity with language model training, evaluation, and inference pipelines.
• Expertise in performance optimization and distributed computing systems.
• Experience with VM/sandboxing/container deployment and large-scale data processing.

Responsibilities

• Identify and remove bottlenecks across the full stack to enable progress toward scientific AGI.
• Develop approaches for long-horizon task completion and complex reasoning challenges.
• Scale research ideas from prototype to production.
• Create benchmarks and evaluation frameworks for model capabilities in scientific workflows.
• Implement distributed training systems and performance optimizations for large-scale model development.

Skills

ML research

Language model training

Evaluation pipelines

Distributed systems

Performance optimization

Problem solving

Collaboration

Education

Bachelor’s degree

Tools

Docker

Kubernetes

VM/sandboxing

Cloud deployment

Job description

Anthropic is seeking a Research Engineer to work end-to-end on bottlenecks toward scientific AGI. You will contribute to language model training, evaluation, and inference, scaling research ideas from prototype to production, while optimizing performance and deploying via VM/sandboxing and cloud infra.

Strong candidates have extensive ML research experience, familiarity with large-scale LMs, and a track record of shipping ML systems for complex multi-step reasoning.