Machine Learning Infrastructure Engineer, Safeguards Research

Anthropic•San Francisco, CA

1d•$350,000 - $500,000•Hybrid

About The Position

Anthropic's Safeguards team builds systems to detect and mitigate misuse of AI models, including individual policy violations and sophisticated, coordinated attacks. A significant part of this work involves lightweight detection methods trained on model internals, enabling cost-effective and scalable identification of harmful behavior. This role focuses on owning the infrastructure that supports this research, including tooling for experiments, training detection methods, and selecting detections for launch. It bridges the gap between research and production, ensuring fast iteration for researchers and reliable results for detection systems as models evolve. The engineer will tackle novel systems problems at scale, building abstractions, pipelines, and tooling to maintain research velocity amidst shifting requirements. The ideal candidate will have a proven ability to solve large-scale systems and data problems and a strong interest in developing deep machine learning expertise.

Requirements

Strong software engineering fundamentals and hands-on coding ability, with proficiency in Python.
Experience building and operating data-intensive or distributed systems in production.
Experience building tooling or infrastructure that other engineers or researchers use as a dependency.
Comfort working across the research-to-deployment pipeline, from exploratory experiments to production systems.
Ability to debug performance and correctness problems across an unfamiliar stack.
Strong written and verbal communication skills, and a collaborative approach to technical decisions.

Nice To Haves

Experience with high-performance, large-scale machine learning systems.
Familiarity with language modeling and transformers, including working with model internals.
Experience with machine learning framework internals, GPU or accelerator programming, or inference optimization.
Experience building experiment tracking, caching layers, or evaluation harnesses for research teams.
Experience with probes, interpretability, or classifier development.
Interest in the misuse risks of AI systems and a desire to work on mitigating them.

Responsibilities

Build and scale the infrastructure and data pipelines behind Safeguards machine learning research.
Own the training, evaluation, and scoring workflows researchers use, with a focus on cutting the time between an idea and a result.
Design tooling and interfaces, including libraries and command line tools, that researchers can use directly without needing to understand the systems underneath.
Build correctness and sanity checking into the stack, so results stay trustworthy as models and workloads evolve.
Take the highest-value research workflows from experiments to reliable, production-grade jobs.
Improve the throughput, cost, and reliability of large-scale inference and scoring workloads.
Partner closely with researchers and engineers across Safeguards to understand their workflows, anticipate how their needs will change, and design for that ahead of time.