Cybersecurity RL Research Engineer: Build Safe, Secure AI

The Briefed; San Francisco (CA); On-site; USD 180,000 - 210,000; Full time

Benefits offered by this job

Competitive compensation and benefits

Optional equity donation matching

Generous vacation and parental leave

Flexible working hours

Office space for collaboration

Job summary

Anthropic is seeking a Research Engineer for the Cybersecurity RL team within Horizons. You will design and build RL-based defenses and contribute to secure coding and vulnerability remediation capabilities for safe AI models.

The role blends research and engineering, requiring cybersecurity expertise paired with experience in training reliable AI systems. Collaboration with researchers, engineers, and cybersecurity specialists will be essential to advance defensive AI capabilities.

Qualifications

• Bachelor's degree or equivalent in a field relevant to the role.
• Experience in cybersecurity research.
• Experience with machine learning.
• Strong software engineering skills.

Responsibilities

• Design and implement RL environments.
• Run experiments and evaluations.
• Deliver your work into production training runs.
• Work with researchers, engineers, and cybersecurity specialists inside and outside Anthropic.
• Develop novel approaches and turn them into working code.
• Help advance model capabilities in secure coding, vulnerability remediation, and defensive cybersecurity.

Skills

Cybersecurity research

Machine learning

Software engineering

RL environments

Experiment design

Education

Bachelor's degree

Job description

Anthropic is seeking a Research Engineer for the Cybersecurity RL team within Horizons. You will design and build RL-based defenses and contribute to secure coding and vulnerability remediation capabilities for safe AI models.

The role blends research and engineering, requiring cybersecurity expertise paired with experience in training reliable AI systems. Collaboration with researchers, engineers, and cybersecurity specialists will be essential to advance defensive AI capabilities.