We are seeking a highly skilled and analytical
AI/ML Model Validation Specialist to join our Risk Management/Data Science team.
Key Responsibilities
- Independent Validation: Conduct rigorous, independent technical validation of AI/ML models throughout their life cycle (pre-development, post-development, and post-implementation).
- Methodology Review: Evaluate the conceptual soundness, mathematical logic, and technical design of AI models.
- Performance Testing: Perform benchmarking, sensitivity analysis, and robustness testing to ensure models behave as expected under various scenarios.
- Data Validation: Assess the appropriateness of data sources, training sets, and data processing techniques to detect biases.
- Risk Assessment: Identify potential weaknesses, biases, and risks (operational, ethical, or financial) associated with model outputs.
- Documentation & Reporting: Author comprehensive validation reports outlining findings, limitations, and recommendations for model enhancement, ensuring transparency for regulators.
- Stakeholder Engagement: Collaborate with developers, data scientists, and business leaders to discuss findings and track remediation efforts.
- Stay Current: Keep up-to-date with emerging AI technologies, validation techniques, and regulatory developments.
- Education: Graduate degree (Masters or PhD preferred) in a quantitative field such as Statistics, Mathematics, Computer Science, Data Science, Physics, or Engineering.
- Experience: 3+ years of hands-on experience in model development, validation, or risk management, specifically within the AI/ML domain.
- Technical Skills:
- Proficiency in Python, R, or similar programming languages.
- Strong understanding of ML frameworks (TensorFlow, PyTorch, scikit-learn).
- Familiarity with cloud platforms (AWS, GCP, or Azure).
- AI Specialization: Experience validating LLMs, GenAI, NLP, or computer vision models is highly desirable.
- Communication: Excellent written and verbal communication skills, with the ability to explain complex technical concepts to non-technical stakeholders.
