Runs the team day to day
In-house, owns the opening
Joined the team recently
From your own Gmail and LinkedIn.
About the role
EY is hiring a full-time senior AI security engineer in Gurugram, India, for someone with 3-6 years of cybersecurity, application security or cloud security experience. The role leads adversarial testing and secures AI and generative AI components across their lifecycle in enterprise environments. It is self-directed, so you'll drive security testing and control implementation for enterprise AI workloads on your own initiative. EY describes its work as spanning assurance, consulting, law, strategy, tax and transactions for clients in more than 150 countries.
What you’ll do
- Lead AI red team exercises covering prompt injection, jailbreaks and data exfiltration.
- Model threats across the model, data and application layers of AI systems.
- Design and build guardrails for large language models and security controls for retrieval-augmented generation.
- Review ML pipelines and model lifecycles for security weaknesses.
- Secure APIs and inference endpoints running in cloud environments.
- Support incident response for AI-specific attacks and exploitation attempts.
- Build security controls into AI and ML pipelines and DevSecOps workflows.
- Work with SOC teams on detection and monitoring of AI workloads.
What they’re looking for
- 3-6 years of experience in cybersecurity, application security or cloud security.
- Hands-on exposure to AI or generative AI security concepts or implementations.
- Experience in vulnerability assessment, red teaming or threat modeling.
- Working knowledge of Azure or AWS.
- Understanding of ML and GenAI concepts such as LLMs, embeddings and pipelines.
- Experience securing applications, APIs and cloud platforms.
- Familiarity with AI attack surfaces across prompt, model and data layers.
- An adversarial mindset for simulating realistic attack scenarios.
- Knowledge of the OWASP Top 10, LLM vulnerabilities and the MITRE ATT&CK framework.
Nice to have
- Hands-on work with AI security tools such as Garak, PyRIT or LLM Guard.
- Exposure to adversarial machine learning tools such as ART or Foolbox.
- Experience integrating security into MLOps or DevSecOps pipelines.
- Certifications in cloud or security domains, such as AZ-500 or CEH.
What’s on offer
- Work on advanced AI security engagements.
- Exposure to global clients.
- Continuous learning in AI-led cyber defense and adversarial testing methods.
Skills
Questions about this role
Where is this senior AI security job based?
In Gurugram, Haryana, India.
How many years of experience are required?
3-6 years in cybersecurity, application security or cloud security.
Is AI security experience needed?
Hands-on exposure to AI or generative AI security concepts or implementations is listed as a requirement.