New opportunity

AI-Driven VM_Cloud Security_GenAI_AI/ML

Infosys Limited · Bangalore, India

About this role

The AI-Driven VM_Cloud Security_GenAI_AI/ML role focuses on securing AI systems and cloud environments. The successful candidate will lead red teaming operations, simulate attacks to evaluate defenses, and address AI-specific attack surfaces including fairness, bias, and anomaly detection. Responsibilities include programming and scripting, conducting security assessments using industry frameworks, and managing client communications. Key Responsibilities: Design and execute security testing strategies for AI systems; utilize penetration testing tools and AI-focused frameworks; identify and mitigate emerging AI security threats such as prompt injection; and provide actionable technical reports for stakeholders. Required Qualifications: 2-15 years of experience; Bachelor's or Master's degree in Engineering, Computer Science, or related fields; strong programming skills in Python; deep understanding of AI/ML model pipelines and security frameworks like OWASP LLM Top 10 and MITRE ATLAS; proficiency in cloud security (AWS, Azure, GCP); and experience leading red teaming operations. Preferred Qualifications: Certifications such as OSCP, GPEN, GWAPT, or CEH; experience managing remote teams; and expertise in vulnerability management tools like Qualys Guard. Soft Skills: Strong analytical, problem-solving, and stakeholder management capabilities are essential for this role.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonCC++JavaJavaScriptAI/MLOWASP LLM Top 10MITRE ATLASNIST AI securityKali LinuxBurpsuiteNmapNessusClever HansARTFool boxAWSAzureGCPPenetration TestingDASTSASTVM ScanExploitation FrameworksGenerative AI

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

AI Red Teaming and Offensive Security
The core of this role is leading red teaming operations specifically for AI systems. You must be able to simulate adversarial attacks, such as prompt injection, to evaluate the robustness of AI defenses. This requires a deep understanding of AI/ML model pipelines and the ability to apply offensive security techniques to identify vulnerabilities in AI-specific attack surfaces.
AI Security Framework Implementation
You are expected to apply industry-standard security frameworks to AI and cloud environments. This includes utilizing the OWASP LLM Top 10, MITRE ATLAS, and NIST AI security guidelines to design and execute comprehensive security testing strategies. Proficiency in translating these frameworks into actionable security assessments is a primary duty.
Technical Reporting and Stakeholder Management
Beyond technical testing, you must bridge the gap between complex security findings and business requirements. The role requires creating clear, actionable technical reports and presentations for stakeholders, as well as managing direct client communications and reporting on project status and issues.
Cloud and Vulnerability Management
The role demands proficiency in securing cloud environments (AWS, Azure, GCP) and managing vulnerability assessments. You must be adept at using standard penetration testing tools (Kali Linux, Burpsuite, Nmap, Nessus) and vulnerability management tools like Qualys Guard to secure both traditional infrastructure and AI-driven systems.

Likely challenges to think through

Evolving AI Threat Landscape
AI security is a rapidly shifting field where traditional penetration testing methods may not suffice. The challenge lies in staying ahead of emerging adversarial tactics like prompt injection or model poisoning. Prepare by demonstrating how you keep your knowledge current with the latest research and how you adapt standard security frameworks to novel AI-specific threats.
Balancing Technical Depth with Client Communication
You will be expected to lead remote teams and communicate directly with clients. The difficulty is translating highly technical red teaming results into business-level risk assessments that non-technical stakeholders can understand. Prepare by practicing how to articulate the business impact of a security vulnerability without losing technical accuracy.
Managing AI-Specific Attack Surfaces
Unlike traditional software, AI systems introduce unique vulnerabilities related to fairness, bias, and data integrity. The challenge is designing tests that account for these non-deterministic behaviors. Prepare by reviewing how to integrate tools like ART or Foolbox into your testing pipeline to systematically evaluate model reliability and security.

Practice your answers

Technical practice

AI Model Attack Surfaces

What to coverExplain how you would identify and mitigate vulnerabilities related to fairness, bias, and anomaly detection in an AI model pipeline. Discuss the trade-offs between model performance and security hardening.

Prompt Injection Mitigation

What to coverDescribe the mechanisms you would implement to defend against prompt injection attacks in a Generative AI application. Address both direct and indirect injection vectors.

Applying MITRE ATLAS

What to coverHow do you utilize the MITRE ATLAS framework to map adversarial tactics against an AI system? Provide an example of how this framework informs your red teaming strategy.

OWASP LLM Top 10 Application

What to coverSelect one vulnerability from the OWASP LLM Top 10 and explain how you would design a security test to detect it. Suggest specific tools or manual techniques for validation.

Red Teaming AI Systems

What to coverWhen leading a red teaming operation for an AI system, how do you simulate realistic adversarial attacks while ensuring the stability of the production environment? Discuss the importance of scoping and rules of engagement.

Cloud Security Integration

What to coverHow does securing an AI model deployed on AWS, Azure, or GCP differ from traditional application security? Focus on the intersection of cloud infrastructure controls and model-specific security.

AI-Focused Security Frameworks

What to coverCompare the use of tools like Clever Hans, ART (Adversarial Robustness Toolbox), and Foolbox in assessing model robustness. When would you choose one over the others?

Vulnerability Management in AI

What to coverHow do you integrate DAST and SAST into an AI/ML development lifecycle? Suggest how to handle findings that are specific to AI models versus traditional software code.

Technical Reporting for Stakeholders

What to coverHow do you translate complex AI security findings into actionable reports for non-technical business stakeholders? Suggest a structure that balances technical depth with business risk impact.

Behavioural practice

Managing Remote Security Teams

What to coverDescribe a situation where you led a remote team through a complex security assessment. How did you ensure clear communication, track progress, and maintain quality across the team?

Stakeholder Communication

What to coverTell me about a time you had to communicate a critical security vulnerability to a client or stakeholder who was resistant to the recommended mitigation. How did you handle the situation?

Problem-Solving Under Pressure

What to coverDescribe a scenario where you identified an unexpected security threat during a penetration test. What was your immediate action, and how did you document the findings for the client?

Cross-Functional Collaboration

What to coverGive an example of how you collaborated with data scientists or developers to remediate a security flaw in an AI model. How did you balance their development speed with security requirements?

Sharpen your knowledge

Topics to prepare

AI Security Frameworks
Review the OWASP LLM Top 10, MITRE ATLAS, and NIST AI security guidelines. Practice mapping these frameworks to specific AI model vulnerabilities, such as prompt injection or data poisoning, to prepare for designing security testing strategies.
AI Model Attack Surfaces
Study the lifecycle of AI/ML model pipelines and identify where fairness, bias, and anomaly detection failures occur. Focus on understanding how adversarial machine learning techniques can be used to manipulate model outputs.
Red Teaming Operations
Research methodologies for leading red teaming exercises. Focus on how to simulate realistic adversarial attacks against AI systems and how to document these findings into actionable technical reports for diverse stakeholders.

Skills to revise

Python for Security Automation
Develop scripts in Python to automate security assessments or interact with AI model APIs. This is essential for both general security tasks and testing AI-specific vulnerabilities.
AI-Focused Security Tooling
Gain hands-on experience with AI-specific security frameworks like Adversarial Robustness Toolbox (ART), CleverHans, or Foolbox. Practice using these to evaluate the robustness of machine learning models against adversarial inputs.
Cloud Security Assessment
Practice conducting security assessments within AWS, Azure, or GCP environments. Focus on securing the infrastructure that hosts AI/ML pipelines, including identity management and network security configurations.

Your preparation checklist

Develop an AI Red Teaming Case Study
Prepare a project walkthrough demonstrating how you identified a vulnerability in an AI/ML pipeline. Focus on your methodology for detecting prompt injection or model bias using frameworks like MITRE ATLAS or the OWASP LLM Top 10, and explain how you documented the remediation steps for stakeholders.
Demonstrate Cloud and AI Security Proficiency
Create a technical summary of your experience securing AI workloads on AWS, Azure, or GCP. Highlight specific configurations or security assessments you performed to protect model endpoints and data pipelines against adversarial attacks.
Prepare Technical Reporting Samples
Draft a template for a security assessment report that translates complex AI-specific threats (e.g., model inversion or poisoning) into actionable insights for non-technical business stakeholders, ensuring the output aligns with the role's requirement for clear communication.
Formulate Strategic Interview Questions
Prepare questions regarding the current maturity of the team's AI security posture, such as: 'How does the team currently integrate MITRE ATLAS into the existing vulnerability management lifecycle?' or 'What are the primary challenges the team faces when balancing AI innovation with security compliance?'
Review Offensive Security Tooling
Refresh your practical knowledge of penetration testing tools mentioned in the job description, such as Kali Linux, Burpsuite, and Nmap, and be prepared to discuss how you have applied these to simulate attacks in a controlled environment.

Availability has not yet been verified. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…