Member of Technical Staff, Vulnerability Researcher

Help strengthen frontier AI systems by finding novel attack paths before they can be exploited in production. In this full-time, remote role, you will operate as an internal adversary across AI agents, cloud infrastructure, developer platforms, and production systems, partnering with engineering teams to make the technology stack more resilient.

• Conduct advanced offensive security research across AWS and GCP infrastructure, production services, internal tools, and AI platforms.
• Design and run realistic adversary simulations against identity systems, cloud control planes, supply chains, and distributed architectures.
• Identify and exploit vulnerabilities in proprietary applications, APIs, infrastructure as code, CI/CD pipelines, and AI-enabled developer workflows.
• Research attack vectors affecting LLMs, AI agents, retrieval systems, MCP integrations, prompt orchestration, and autonomous workflows.
• Reverse engineer critical services to uncover architectural weaknesses, new exploitation techniques, and previously unknown vulnerability classes.
• Model insider and advanced persistent threat scenarios, including privilege escalation, lateral movement, and defense evasion in modern enterprise environments.
• Build offensive tooling, automation frameworks, fuzzers, and proof-of-concept exploits to accelerate vulnerability discovery.
• Work with infrastructure, product, and AI engineering teams to validate remediations, improve secure-by-design practices, and validate security controls.
• Publish internal research, document attack methods, and help guide long-term security strategy.

• Proven experience in offensive security, vulnerability research, red teaming, or exploit development.
• Deep cloud security knowledge across AWS and GCP, including IAM, networking, Kubernetes, containers, and identity systems.
• Strong application security, code auditing, reverse engineering, and vulnerability discovery experience.
• Experience finding weaknesses in software supply chains, CI/CD systems, APIs, and infrastructure automation.
• Proficiency in Python, Go, Rust, C/C++, or similar languages used for security research and offensive tooling.
• Strong understanding of operating system internals, networking, authentication protocols, and modern security architectures.
• Ability to independently investigate ambiguous problems and develop novel attack techniques.
• Excellent written communication skills and the ability to explain complex vulnerabilities to engineering teams.
• Preferred experience securing AI and LLM systems, including agentic systems, prompt injection, tool abuse, indirect prompt attacks, jailbreaks, model manipulation, RAG security, or autonomous workflows.
• Preferred experience discovering zero-day vulnerabilities, developing exploits, or contributing to offensive security research.

• Full-time, remote position.

• Annual compensation range: $240, 000 to $400, 000.