• Write domain-specific CBRN prompts, including adversarial prompts designed to test model vulnerabilities and resistance to generating dangerous outputs • Evaluate AI-generated responses for accuracy, scientific validity, and regulatory compliance • Identify and flag content that is incorrect, harmful, misleading, or ethically concerning • Provide constructive feedback to improve prompt design and content generation • Stay updated with current research, regulations, and best practices within your area of expertise

• Conduct sophisticated and comprehensive simulated attacks on generative AI models and their operating environments to uncover vulnerabilities. • Evaluate the security posture of AI models and infrastructure, identifying weaknesses and potential threats. • Perform thorough risk analysis to determine the impact of identified vulnerabilities and prioritize mitigation efforts. • Collaborate with development and security teams to develop effective strategies to mitigate identified risks and enhance model resilience. • Stay abreast of the latest trends and developments in AI security, ethical hacking, and cyber threats. • Maintain detailed documentation of all red team activities, findings, and recommendations. • Prepare and present reports to senior management and relevant stakeholders.

• Collaborate with business stakeholders to identify opportunities where AI agents can improve efficiency, accuracy, and customer experience. • Gather, analyze, and document functional and non-functional requirements for AI agent use cases. • Translate business workflows and policies into AI agent logic, decision trees, prompts, and automation rules. • Analyze AI agent performance using qualitative feedback and quantitative metrics (e.g., accuracy, resolution rates, escalation rates, latency). • Monitor AI agent behavior to identify gaps, bias, errors, or unintended outcomes and recommend corrective actions. • Partner with engineering and data science teams to support AI agent development, testing, deployment, and iteration. • Define success metrics, KPIs, and reporting dashboards for AI agent effectiveness and business impact. • Ensure AI agents comply with regulatory, risk, and responsible gaming requirements. • Support UAT, validation, and post-launch monitoring of AI agent releases. • Document AI agent workflows, decision logic, escalation paths, and operational guidelines. • Work closely with Customer Operations, Fraud, Risk, and Compliance teams to align AI behavior with real-world processes. • Facilitate feedback loops between end users and technical teams to drive continuous improvement. • Stay current on emerging AI, automation, and agent-based technologies and recommend enhancements. • Proactively identify optimization opportunities to expand AI agent capabilities across the organization.

Future Opportunities: Arizona-Licensed Special Education & Related Service Providers (Remote, Contract) Location: Arizona (Remote) Position Type: Contract Student Population: Homeschooled and Private School Students Schedule: As-needed / Flexible MIYO Health is launching a new initiative to build a talent pool of Arizona-licensed special education and related service providers in anticipation of future opportunities. While roles may not be immediately available, we are actively seeking qualified professionals who are interested in part-time, remote contract work should services be requested by families or partner organizations. Who We’re Looking For We are seeking Arizona-licensed providers in the following areas: - Speech-Language Pathology (SLP) - Occupational Therapy (OT) - Physical Therapy (PT) - Special Education Teachers - Teachers of the Visually Impaired (TVI) - School Counselors/School Social Workers - Other related service providers supporting special education needs About the Services - Services are delivered remotely to students who are homeschooled and/or attending private schools - Support is provided on an as-needed basis, depending on student demand - Providers work independently while collaborating with MIYO Health’s internal team as appropriate