Machine Learning Research Engineer – Safeguards

About the role

Anthropic is looking for a Machine Learning Research Engineer to join its Safeguards team. You will help detect and mitigate misuse of AI systems, ensuring they remain safe and beneficial as capabilities grow. The role bridges research and production, focusing on real‑world safety challenges.

Key responsibilities

• Develop classifiers that detect misuse and anomalous behavior at scale, including synthetic data pipelines for training.
• Build monitoring systems for multi‑exchange harms such as coordinated cyber‑attacks and influence operations.
• Design threat models, test environments, and mitigations for agentic risks and prompt‑injection attacks.
• Conduct research on automated red‑teaming, adversarial robustness, and other methods to uncover misuse.

Required profile

• 4+ years of experience in ML engineering, research engineering, or applied research.
• Proficiency in Python and experience building end‑to‑end ML systems.
• Comfort with the full research‑to‑deployment pipeline, from experiments to production.
• Strong interest in AI safety and mitigating misuse risks.
• Excellent communication skills for explaining technical concepts to non‑technical stakeholders.

Required skills

• Python
• Language modeling and transformers
• Classifier and anomaly‑detection development
• Adversarial machine learning / red‑teaming
• Interpretability techniques
• Reinforcement learning
• High‑performance, large‑scale ML systems

What we offer

• Competitive annual salary ranging from $350,000 to $500,000 USD