AI Safety and Responsible AI Lead - Jersey City, NJ (Onsite)

[Sid K]

Hi Folks,

My client is looking for AI Safety and Responsible AI Lead for 12 Month Contract role based in Jersey City, NJ (Onsite)

 

Position: AI Safety and Responsible AI Lead

Location:  Jersey City, NJ (Onsite)

Duration: 12 Month Contract

Note: The client is looking for a candidate with 15+ years of experience. 

Primary ownership

Responsible AI policy, control framework, risk taxonomy, governance workflows, and production-readiness criteria.

AI risk assessments, impact assessments, safety evaluations, and post-production monitoring standards.

Cross-functional alignment across engineering, product, legal, compliance, model risk, audit, cybersecurity, and data governance.

 

Key responsibilities

Define Responsible AI standards, policies, procedures, risk-classification methods, and operating models for AI and GenAI initiatives.

Establish governance processes for use-case intake, risk assessment, model review, approval workflows, deployment readiness, and ongoing monitoring.

Develop safety and evaluation frameworks covering fairness, bias, explainability, transparency, robustness, privacy, hallucination, harmful outputs, and human oversight.

Define guardrail requirements for LLMs, RAG systems, agentic workflows, and high-risk AI applications.

Partner with model risk, legal, compliance, data governance, cybersecurity, and audit teams to align AI controls with enterprise expectations.

Lead AI impact assessments, risk reviews, control assessments, readiness reviews, and remediation planning.

Establish metrics and monitoring for bias indicators, safety violations, explainability gaps, harmful outputs, user feedback, and behavior drift.

 

Must-have candidate profile

10+ years of Technology ethics & safety

Deep understanding of Responsible AI, AI ethics, model governance, model risk, explainability, fairness, privacy, safety, and enterprise risk management.

Experience implementing AI governance or Responsible AI controls in production or enterprise environments.

Understanding of LLM-specific risks such as hallucination, bias, toxicity, prompt injection, data leakage, overreliance, and unsafe automation.

Ability to translate policy and regulatory expectations into practical product and engineering controls.

Thanks 
Sid