AI Risk Architect: Safeguards, Fairness & Reliability

R. L. Polk Mobility LLC; London; On-site; CAD 120,000 - 180,000; Full time

Job summary

R. L. Polk Mobility LLC is seeking a on-site AI risk governance professional to maintain risk taxonomy, score AI risk across safety, fairness, reliability, and explainability, and support risk documentation templates for project teams.

Responsibilities include developing rubrics for bias, fairness, explainability, drift tolerance, and harmful outputs, plus coordinating with Engineering and Cybersecurity on risk audits and red-teaming activities.

Qualifications

• Experience in AI/ML risk management or related analytical function.
• Familiarity with AI-specific frameworks and legislation (NIST AI RMF, ISO 42001 & 23053, EU AI Act).
• Hands-on experience with model evaluation, bias and fairness testing, or AI red-teaming is a plus.

Responsibilities

• Maintain the organization's AI risk taxonomy and assessment methodology covering safety, fairness, reliability, explainability, and harmful output risks.
• Develop risk evaluation rubrics for AI dimensions (bias testing, fairness metrics, explainability, drift, harmful output).
• Develop and maintain templates for risk documentation (model cards, risk assessments, impact analyses, risk scorecards).
• Support project teams in applying risk framework and quality gates; escalate high-risk deployments for review.
• Coordinate red-teaming and adversarial testing with Cybersecurity focusing on safety, bias, and harms.

Skills

AI risk management

Model risk management

Risk assessment

Bias testing

Explainability

Fairness metrics

Job description

R. L. Polk Mobility LLC is seeking a on-site AI risk governance professional to maintain risk taxonomy, score AI risk across safety, fairness, reliability, and explainability, and support risk documentation templates for project teams.

Responsibilities include developing rubrics for bias, fairness, explainability, drift tolerance, and harmful outputs, plus coordinating with Engineering and Cybersecurity on risk audits and red-teaming activities.