Machine Learning Engineer – Secure AI & Anti-Abuse

About the role

Proton’s Mail Delivery, Spam and Anti‑Abuse (MSA) team is looking for a Machine Learning Engineer to help build next‑generation security systems. You will work on real‑time and asynchronous decision‑making pipelines that protect millions of users from spam, abuse, account compromise and DDoS attacks.

Key responsibilities

• Design, implement and deploy scalable ML models using modern MLOps practices, including real‑time inference, monitoring and automated retraining.
• Architect agentic AI components that can autonomously detect threats, enforce adaptive policies and initiate self‑healing security responses.
• Develop tooling for model debugging, explainability and rapid iteration.
• Optimize ML workflows for distributed processing of large‑scale data streams.
• Collaborate with security analysts, backend/frontend engineers and support teams to align solutions with user needs.
• Research and prototype advanced techniques such as adversarial robustness, graph‑based anomaly detection and privacy‑preserving ML.

Required profile

• Strong interest in security, privacy and anti‑abuse problems.
• Experience working in fast‑paced, cross‑functional engineering teams.
• Ability to translate research ideas into production‑ready systems.
• Excellent problem‑solving skills and a bias toward impact.

Required skills

• Machine learning model development and deployment.
• MLOps pipelines for real‑time inference and monitoring.
• Distributed data processing concepts.
• Understanding of security‑related ML techniques (e.g., adversarial robustness, graph anomaly detection).