Responsible AI Engineer

Responsible AI Engineer

About the role

As a Responsible AI Engineer, you will help drive trust, safety, and governance capabilities for agentic AI systems used at enterprise scale. You will be part of the AI Market Unit (FDE Pod) and partner closely with AI engineers, architects, risk teams, and client stakeholders to ensure every AI solution is safe, compliant, and ready for production.

In this role, you will:

• 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
• Drive red-teaming and adversarial testing efforts to uncover vulnerabilities such as jailbreaks, prompt injections, and trust boundary violations prior to production rollout
• Develop and operationalize AI governance and compliance frameworks, mapping regulations (e.g., EU AI Act, NIST AI RMF) into enforceable engineering controls
• Create and deploy fairness, bias, and model evaluation pipelines, including hallucination detection, groundedness validation, and subgroup performance analysis
• Implement observability and auditability for AI systems using structured logging, audit trails, governance metrics, and incident response processes

Work model

At Cognizant, we aim to offer flexibility wherever possible, and we support a healthy work-life balance through our wellbeing programs. Depending on the needs of this role, this is an onsite position requiring 5 days a week in a Cognizant office in Chennai, Bangalore, or Hyderabad.

The working arrangements described for this role reflect the date of posting. These may evolve depending on the project you are assigned to, along with business and client requirements. We will clearly communicate any changes to role expectations.

What you need to have to be considered

• Experience in software engineering or ML engineering, including practical experience in AI safety, governance, or trust engineering
• Demonstrated 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
• Capability to partner with security, legal, and risk stakeholders, converting technical AI risks into actionable guidance for business leaders
• Outstanding communication and executive-level presentation skills, including experience presenting to senior client stakeholders