Role: AI Offensive Security Researcher - Red Team, VAPT & Agentic Security
Location: Bangalore, Hybrid
About the role...
We are looking for a AI offensive Security researcher / Senior AI Security Engineer/Architect with MLSecOps experience to lead the next generation of AI-native security at o9.
You will own two parallel missions: securing o9’s GenAI platform and agentic AI ecosystem from emerging threats, and building AI/ML-driven capabilities that transform how our security operations detect, predict, and respond to adversarial activity across 500+ customer environments.
Our security platform ingests 22 billion events per month across 176,000+ hosts, 266 Kubernetes clusters, and 3,042 cloud accounts. The intelligence hidden in that telemetry remains largely unmined. Simultaneously, o9’s GenAI platform is scaling rapidly - production AI agents, RAG pipelines, MCP tool integrations, and autonomous planning workflows that introduce a fundamentally new class of security risk. This role exists to address both.
What you’ll do:
- Design and execute AI-specific VAPT engagements - prompt injection chains, RAG pipeline poisoning, agent identity spoofing, and tool-layer exploitation across MCP and A2A boundaries
- Lead structured AI red team exercises and scenario-based tabletops, delivering CVSS - scored, MITRE ATLAS mapped findings with actionable remediation guidance
- Build and maintain AI security automation tooling: prompt fuzzing harnesses, guardrail regression test suites, and detection rule generators fed by red team TTP libraries
- Contribute to threat modeling for agentic systems -mapping trust boundaries, privilege escalation paths, and blast radius across autonomous agent clusters
- Feed red team findings back into detection engineering as quality inputs, closing the loop between offensive discovery and SOC coverage gaps
What you'll have:
- 4-12+ years in application security, AI/ML security, or offensive security research with at least one demonstrable AI-specific assessment or published finding
- Hands-on offensive depth across LLM attack surfaces: jailbreaks, multi-modal injection, agentic control gaps, and indirect prompt injection
- Working knowledge of MLOps pipelines, ML engineering concepts, or cloud security as contextual grounding
- CFP submissions, responsible disclosure history, or published AI security research is a strong differentiator
What we’ll do for you:
- Flat organization: With a very strong entrepreneurial culture (and no corporate politics).
- Great people and unlimited fun at work.
- Possibility to really make a difference in a scale-up environment.
- Support network: Work with a team you can learn from every day.
- Diversity: We pride ourselves on our international working environment.
- DavOs Insight (Turning VUCA into value with Neuro-Symbolic AI): https://l1nk.dev/ngQoe
- Work-Life Balance: https://youtu.be/IHSZeUPATBA?feature=shared
- Feel part of A team: https://youtu.be/QbjtgaCyhes?feature=shared
How the process works...
- Setup & Create your profile in Workday to track the status - Link to apply (JR102834). If already done, kindly ignore!
- Respond with your interest to us.
- We’ll contact you either via video call or phone call - whatever you prefer, with the further schedule status.
- During the interview phase, you will meet with the technical panel for 60 minutes. We will contact you after the interview to let you know if we’d like to progress your application.
- There will be 2 rounds of technical discussion followed by a Managerial round.
- The 1st round will be a virtual interview. The 2nd and HM rounds will be in-person interviews.
- We will let you know if you’re the successful candidate.
Good luck!