AI Model & Dataset Evaluation
Define and own evaluation frameworks for ML, Computer Vision, LLM, RAG, and Agentic AI systems.
Establish statistically rigorous model validation methodologies.
Assess dataset quality, provenance, coverage, and labeling integrity.
Conduct significance testing before approving model improvements.
AI Security & Red Teaming
Lead adversarial testing and red-team exercises.
Evaluate resilience against data poisoning, evasion attacks, prompt injection, jailbreaks, and tool abuse.
Develop security benchmarks and release-gating criteria for AI deployments.
Manage vulnerability identification and remediation tracking.
AIOps & Production Reliability
Design drift monitoring and operational observability frameworks.
Build shadow deployment and canary rollout strategies.
Architect high-throughput, low-latency AI inference pipelines.
Develop confidence calibration and model reliability monitoring.
Governance & Compliance
Ensure reproducibility, auditability,



and certification readiness.
Maintain AI lifecycle documentation and evaluation records.
Enforce secure data handling and access-control practices.
Technical Leadership
Mentor ML engineers, security researchers, and AI practitioners.
Establish engineering standards for secure AI development.
Advise leadership on AI risk, deployment readiness, and technology selection.

Required Skills
6 - 10 years of experience in Machine Learning, AI Engineering, Security Engineering, or related domains.
Solid expertise in:
Statistical testing and model validation
Classification and detection evaluation
Adversarial Machine Learning
LLM, RAG, and Agentic AI evaluation
MLOps and production AI systems
Drift detection and monitoring
Secure deployment practices in high-security environments

📌 AIOps (Bengaluru)
🏢 KPMG India
📍 Bengaluru