Software Engineer, Infrastructure, Interpretability

Overview

Lead early infrastructure for Interpretability, building secure-by-default, privacy-focused research access and reliable audit pipelines for frontier model work.

What you'll do

• Design, build, and own shared Interpretability infrastructure (research environments, data systems, compute tooling).
• Lead cross-team efforts with agentic engineering, security, compute, and storage platform teams.
• Discover and resolve major organization-wide developer experience issues.
• Help move interpretability methods from research code to dependable audit pipelines.

What you'll need

• Be highly proficient in at least one programming language and productive with Python.
• Have significant experience building and operating secure and scalable software infrastructure.
• Have strong cross-functional communication skills.
• Be extremely curious about unfamiliar domains and interpretability research.
• Prioritize impactful work and be comfortable operating with ambiguity.

Details

• Based in the San Francisco office; remote may be considered case-by-case for exceptional candidates.