AI Red Teamer

This is a fully remote position, open to applicants in India.

📋 Description

• Design and implement red team initiatives targeting AI/ML systems, including prompt injection, jailbreaks, model extraction, adversarial inputs, and training data poisoning.

• Develop and sustain adversarial test suites for our AI penetration testing agent.

• Emulate real-world attacker tactics, techniques, and procedures (TTPs) to assess the decision-making capabilities and guardrails of the agent.

• Investigate new attack vectors aimed at LLM-based systems and agentic workflows.

• Partner with engineering teams to strengthen the AI agent against evasion and manipulation attempts.

• Record findings with clear severity assessments, reproducible proof of concepts (POCs), and remediation recommendations.

• Contribute to Simbian's red team playbooks and processes for threat modeling.

⛳️ Requirements

• 3-6 years of experience in offensive security, red teaming, or penetration testing.

• Practical experience in at least one area: LLM security, adversarial machine learning, or testing AI systems.

• Solid foundation in web application, API, and network penetration testing (OWASP, MITRE ATT&CK).

• Proficient in scripting: Python is essential; familiarity with LangChain, LlamaIndex, or similar AI frameworks is a significant advantage.

• Working knowledge of prompt injection, jailbreaking methods, and OWASP Top 10 for LLMs.

• CEH, OSCP, or equivalent certifications are preferred but not mandatory.

• Excellent written communication skills: your POCs should be comprehensible to both engineers and executives.

🏝️ Benefits

• Engage with a problem that many security teams have yet to define.

• Make a direct impact on the product; your insights will influence our development.

• Enjoy a remote-first, asynchronous-friendly work culture.

• Receive competitive compensation and equity at an early-stage AI security firm.

• Collaborate with professionals who have successfully launched security products at scale.