Join Anthropic’s Cybersecurity RL team within Horizons to combine reinforcement learning research with defensive cybersecurity capabilities.

Safely expand model capabilities for secure coding, vulnerability remediation, and other defensive cybersecurity tasks
Develop new approaches and translate them into working code
Design and implement reinforcement learning (RL) environments
Run experiments and perform evaluations
Move research outputs into production training runs
Collaborate with researchers, engineers, and cybersecurity specialists across and outside Anthropic

Cybersecurity domain expertise, with interest or experience in training safe AI models
Experience in cybersecurity research
Experience with machine learning
Strong software engineering skills
Ability to balance research exploration with engineering implementation
Passion for AI’s potential and commitment to developing safe and beneficial systems

Professional experience in security engineering, fuzzing, detection and response, or other applied defensive work
Experience participating in or building CTF competitions and cyber ranges
Academic research experience in cybersecurity
Familiarity with RL techniques and environments
Familiarity with LLM training methodologies

Bachelor’s degree or an equivalent combination of education, training, and/or experience

Location: San Francisco, CA (hybrid)
Hybrid policy: expected to be in an office at least 25% of the time; some roles may require more
Visa sponsorship: Anthropic sponsors visas, but not for every role and candidate
Field of study: a field relevant to the role as demonstrated through coursework, training, or professional experience
Years of experience: will correlate with internal job level requirements
Fair chance: for roles performing work in California, qualified applicants with arrest or conviction records may be considered consistent with applicable state and local laws
Encouragement to apply: apply even if you do not meet every qualification
Safety: recruiters only contact you from @anthropic.com; legitimate recruiters will not ask for money, fees, or banking information before your first day; confirmed openings are available at anthropic.com/careers

Horizons leads Anthropic’s reinforcement learning (RL) research and development
Contributed to every Claude release, with significant impact on autonomy, coding, and reasoning capabilities

Focuses on high-impact AI research as big science
Operates as a cohesive team on a few large-scale research efforts
Values impact toward long-term goals of steerable, trustworthy AI
Treats AI research as empirical science
Highly collaborative, with frequent research discussions and emphasis on communication
Directions are explained through recent research; prior directions include GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences

Public benefit corporation headquartered in San Francisco
AI usage guidance: policy on using AI in the application process is provided via Anthropic careers

Research Engineer, Cybersecurity RL

Job Description