Senior Research Scientist, AI for Cybersecurity

Description

The team is looking for a senior scientist to lead research on AI models and agents for cybersecurity and to apply them to hard security problems.

Work covers improving security reasoning, code and system understanding, planning, tool use, and long-horizon investigation, along with how agents collaborate with human defenders. It also covers applying these capabilities to vulnerability discovery, secure development, threat and anomaly detection, malware and attack analysis, incident investigation, security validation, and automated defense.

You will advance the underlying models, agent architectures, data, training and post-training methods, and evaluations needed for effective cybersecurity. The work must meet the demands of real security environments: high precision, robustness against adaptive adversaries, explainability, privacy, secure deployment, and effective human oversight.

You will link foundational AI and security research to practical tools and workflows that improve prevention, detection, investigation, and response. The company builds accelerated computing and AI systems. The role includes setting research direction, influencing future platforms, and turning foundational advances into practical capabilities with research, product, and engineering teams.

Responsibilities

• Set a high-impact research agenda and lead projects on AI models and agents for cybersecurity and their use on important security problems.
• Improve model and agent abilities in security reasoning, code and system understanding, planning, tool use, memory, long-horizon investigation, and working with human defenders, through work on architectures, data, training, post-training, and evaluation.
• Build AI-based capabilities for vulnerability discovery, secure development, threat detection, malware analysis, incident investigation, security validation, and automated defense that hold up against adversarial manipulation, uncertainty, distribution shift, and incomplete or noisy operational data.
• Build prototypes and test them in realistic environments, measuring detection quality, false positives, time to insight, analyst effectiveness, security risk, and operational scalability.
• Work across AI, security, research, engineering, and product teams, publish original research, mentor researchers and interns, and help move advances into deployed capabilities.

Requirements

• A Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, or a related field, or equivalent experience.
• 5+ years of relevant experience, with a proven research record shown through influential publications, open-source systems, patents, technology transfer, or substantial research collaborations.
• Deep expertise in at least one area such as agentic AI, model training, reasoning and tool use, machine learning for security, vulnerability research, program analysis, software security, intrusion detection, malware analysis, or security operations.
• Experience defining and leading ambiguous, multi-disciplinary research projects and mentoring others technically.
• Strong experimentation and prototyping skills in Python and relevant AI/ML frameworks, including training, adapting, or evaluating models or agents. Experience with systems software, security tooling, or large-scale data systems is required.
• Excellent written and verbal communication, and the ability to work well with both AI researchers and security practitioners.
• You are a creative, collaborative, self-directed researcher who can advance AI agents for cybersecurity and turn new ideas into high-impact research and technology.

Conditions

Base salary depends on location, experience, and pay of employees in similar roles. The base salary range is 224,000 USD - 356,500 USD for Level 5, and 272,000 USD - 431,250 USD for Level 6.

Applications are accepted at least until October 26, 2026.

This posting is for an existing vacancy.

The company uses AI tools in its recruiting process.

The company is an equal opportunity employer and does not discriminate on protected characteristics.