AI Security Researcher Expert

Microsoft • Herzliya,

7d

About The Position

Microsoft is seeking a Principal Security Researcher to lead research for key agents within Project Perception, part of Microsoft’s broader agentic security evolution. The role spans the research lifecycle: framing high-impact investigation problems, applying security expertise to agent design and evaluation, and partnering with data science, engineering, product, and other teams to move capabilities into production and customer-facing experiences. The Microsoft Defender Research team develops autonomous protection to detect, investigate, and disrupt sophisticated attacks across endpoints, identities, cloud applications, and cloud infrastructure, using Microsoft’s global reach and security telemetry. The successful candidate will help transform security operations and shape how defenders investigate and stop attacks at scale.

The researcher will set strategy, use evidence to influence roadmaps, and communicate findings and recommendations to senior stakeholders. They will analyze large-scale security data, improve agent quality and effectiveness, contribute to Microsoft Defender’s protection knowledge layer, and apply AI tools and workflows to accelerate research while validating AI outputs. The role also includes mentoring researchers and raising the team’s technical bar.

Candidates must be able to meet Microsoft, customer, and/or government security screening requirements, including passing the Microsoft Background Check and Microsoft Cloud Background Check upon hire or transfer and every two years thereafter. The position will be open for a minimum of five days, with applications accepted on an ongoing basis until filled.

Requirements

• At least 10 years of experience in security research or a closely related cybersecurity research role.
• Strong, evidence-backed adoption of AI in technical work, including regular use and development of AI skills, coding agents, agentic tools, or AI-assisted workflows, and evaluation and validation of outputs.
• Deep expertise in security research and modern attacker tradecraft, with the ability to conduct complex investigations and translate security knowledge into reliable agent and protection outcomes.
• Experience leading technical work from research and design through production delivery, partnering with engineering and product teams, and influencing roadmaps and senior stakeholders across organizational boundaries.
• Proficiency in programming such as Python or C#, and large-scale query languages such as KQL or SQL, with experience analyzing security data and developing or validating research and protection capabilities.
• Demonstrated technical leadership through mentoring researchers, raising the quality bar, and communicating complex findings clearly.
• Ability to meet Microsoft, customer, and/or government security screening requirements, including required Microsoft background checks.

Nice To Haves

• Bachelor’s, master’s, or doctoral degree in Computer Science, Computer Security, Software Engineering, or a related field; equivalent practical experience is also valued.

Responsibilities

• Lead security research for agents within Project Perception, from problem definition and research methodology through agent design, rigorous evaluation, production readiness, and customer-facing experiences.
• Set research strategy and align partner teams across the organization; translate research priorities into requirements and experiments, use evidence to influence roadmap decisions, and communicate strategy, findings, and recommendations to senior stakeholders.
• Partner with engineering, data science, and product teams to define success criteria, validate quality and safety, establish production readiness, measure customer impact, and iterate based on real-world performance.
• Analyze large-scale data and conduct investigations to improve Project Perception agent quality and effectiveness, building on and contributing to Microsoft Defender’s protection knowledge layer.
• Apply and champion AI in security research using current models, agentic architectures, coding agents, and tooling to accelerate triage, investigation, and detection; validate AI outputs and develop reusable workflows.
• Mentor and grow researchers and elevate the team’s technical bar.

What This Job Offers

• Job Type: Full-time
• Career Level: Principal
• Education Level: No Education Listed