Overwatch - Guardrails Engineer

Job DescriptionRole:Guardrails Engineer | Responsible AI | AI SecurityWe're looking for a Guardrails Engineer to help build and validate the controls that keep AI models and agents safe, secure, compliant, and production-ready.In this role, you'll work at the intersection of Responsible AI, adversarial testing, security engineering, and model-risk management—developing automated ways to challenge AI systems and generate reliable evidence that they meet organizational and regulatory requirements.What You'll Do• Design and automate adversarial and red-team testing for AI models and agents• Develop and validate content, safety, security, and policy guardrails• Build reusable Python-based test harnesses and automation frameworks• Test AI systems against abuse, policy violations, security risks, and unexpected behaviors• Support model and agent enablement gates by validating required controls before production deployment• Automate the collection, validation, and packaging of control and test evidence• Partner with AI/ML, security, risk, compliance, and engineering teams• Help establish repeatable testing standards for regulated AI delivery• Identify weaknesses in AI controls and recommend improvementsWhat We're Looking ForRequired/Strong Experience:?? Responsible AI / AI Governance?? Adversarial testing and AI red teaming?? Content and policy guardrails?? AI/ML security testing?? Model-risk controls and validation?? Python automation and test-harness development?? Evidence automation and reporting?? Experience delivering technology in regulated environmentsIdeal CandidateYou're someone who enjoys breaking AI systems before attackers or real users do. You can think like an adversary, engineer robust automated tests, understand risk and controls, and translate technical test results into evidence that engineering, risk, and compliance teams can trust.If you're passionate about making AI safer, more secure, and production-ready, we'd love to hear from you.