Partner with cross-functional operational stakeholders to evaluate technical feasibility, enforce data privacy and security frameworks, and align systems with regulatory governance standards.
Integrate enterprise applications through authenticated APIs and MCP where appropriate, selecting deterministic workflows, agents or conventional ML based on the problem.
Design evaluations that reflect how products are used. Develop representative datasets, quality rubrics and test cases with domain experts, including challenging scenarios and important failure modes
- Lead end-to-end resource delivery for Edge POP deployments based on network expansion plans and business requirements, ensuring on-time and high-quality execution.
You will run governance (steering committees, executive status reports), JIRA/Kanban stand-ups and post-implementation reviews, working with Regional PMO and Head Office.
Configure and tune Bedrock Guardrails (content filters, denied topics, PII filters, contextual grounding checks) and apply the ApplyGuardrail API in agent and application flows.
Own the selection, evaluation and cost/performance optimisation of foundation models available on Amazon Bedrock (e.g. Anthropic Claude, Amazon Nova, Meta Llama, Mistral).
Serve as a trusted technical advisor to strategic customers, regulators, standards organizations, and industry forums. Represent UL Solutions through technical committees, publications, conferences, and regulatory engagement.
Hands-on Technical Leadership: Stay technically involved and step in when needed to write code, debug issues, review architecture and resolve complex engineering problems.
Demonstrated ability to evaluate and test AI systems and identify failure modes, risks, and control gaps, including model performance, transparency, traceability, explainability, reliability, robustness, human-in-the-loop controls, fairness and bias, privacy, security, accountability, and operational resilience.