Principal Security Researcher
Core
Conduct hands-on vulnerability research, build security research tooling, and lead the MDASH security research team to improve AI agent security and vulnerability detection capabilities.
Role type
Principal Security Researcher (AI Security & Vulnerability Research)
Builds
MDASH agents, tools, model configurations, analysis methods, fuzzing harnesses, datasets, graders, and automation.
Domain
Cyber Security, AI Security, Software Vulnerability Research
Required skills
vulnerability discovery and validation, exploitability assessment, fuzzing, static/dynamic analysis, reverse engineering, symbolic execution, taint analysis, exploit development, security tooling development, AI agent evaluation, prompt orchestration, reinforcement learning for security, secure software development lifecycle, cloud-native security, C/C++/Java/Python/JavaScript security, security benchmarking, CVE disclosure, security community contributions.
Preferred skills
experience with large-scale computing, threat modeling, anomaly detection, memory corruption, injection, authentication/authorization, cryptography, deserialization, path traversal, SSRF, business-logic flaws.
Technologies
C/C++, C#, Java, JavaScript, TypeScript, Python, SARIF, fuzzing frameworks, LLMs, reinforcement learning frameworks.
Responsibilities
Discover and validate vulnerabilities across multiple vulnerability classes and codebase architectures; translate research insights into implemented improvements for MDASH agents and tools; author trusted ground truth spanning vulnerability evidence and attack paths; drive eval-driven development by running evaluations and creating adversarial cases; build research prototypes and automation to accelerate capability improvement; provide technical leadership by shaping research direction and mentoring researchers; collaborate across research, engineering, and product teams to deliver security improvements.
Seniority
Principal, hands-on IC with strategic leadership