Cognizant

Responsible AI Engineer

Cognizant

Chennai•On-site•1-3 Yrs•2 hours ago

Responsible AI Engineer About the role

Design and implement runtime guardrails and safety architectures for LLM and agentic AI systems, including input/output controls, prompt injection detection, and policy enforcement mechanisms
Lead red-teaming and adversarial testing initiatives to identify vulnerabilities such as jailbreaks, prompt injections, and trust boundary violations before production deployment
Build and operationalize AI governance and compliance frameworks, translating regulations (e.g., EU AI Act, NIST AI RMF) into enforceable engineering controls
Develop and deploy fairness, bias, and model evaluation pipelines, including hallucination detection, groundedness validation, and subgroup performance analysis
Establish observability and auditability for AI systems through structured logging, audit trails, governance metrics, and incident response processes

Proven experience building and deploying LLM or agentic AI systems in enterprise or regulated environments
Strong expertise in at least two of the following: guardrails engineering, red teaming, bias/fairness evaluation, AI regulatory compliance
Working knowledge of AI regulatory frameworks such as EU AI Act, NIST AI RMF, ISO/IEC 42001, or sector-specific compliance standards
Strong Python programming skills, with experience building evaluation pipelines, observability systems, or AI governance tooling
Ability to collaborate with security, legal, and risk stakeholders, translating technical AI risks into actionable insights for business leaders
Excellent communication and executive-level presentation skills, with experience presenting to senior client stakeholders
Hands-on experience conducting red-team assessments on production AI systems and remediating critical vulnerabilities
Ability to influence AI system design decisions by balancing technical feasibility with regulatory and risk considerations
Exposure to enterprise AI tooling such as NeMo Guardrails, Giskard, Arize, LangChain/LangGraph, or similar ecosystems

Browse Jobs