What Youll Do:
Drive AI Security Research: Investigate vulnerabilities in AI models, agentic platforms, and infrastructure- covering risks like prompt injection, adversarial manipulation, and model abuse.
Build Proof-of-Concepts: Develop technical PoCs that demonstrate real-world security threats and defenses, supporting both product validation and thought leadership.
Research new Prompt injection techniques: Implement novel techniques of adversarial attacks on AI systems in our automated red team module to evaluate resilience and inform customer-focused security offerings.
Track Competitors & Trends: Research the broader ecosystem of AI and cybersecurity to identify competitor capabilities, market shifts, and innovation opportunities.
Collaborate Across Teams: Partner with engineering and product to translate research into practical protections, product features, and marketing content.
Requirements for Success:
5+ years in security research or offensive engineering, with a focus on high-value targets
Strong background in application and AI security (prompt injection, adversarial ML, agent security).
Experience in red teaming or offensive security, including AI/ML pipelines, APIs, and infrastructure.
Proven ability to design and communicate PoCs for both technical and non-technical audiences.
Familiarity with bug bounty programs or military-grade research is a strong plus.
Curious, detail-oriented mindset with the ability to balance multiple research initiatives.
Experience with research for marketing, including blogs, thought leadership, conferneces, published researches- advantage.
