Senior Agentic AI Engineer for Secure Systems

Socket.dev; United States; On-site; USD 124,000 - 186,000; Full time

Job summary

University of Louisville seeks a Senior AI Engineer to lead the design and delivery of secure, end-to-end multi-agent AI systems. You will architect orchestration across distributed agents with auditable delegation, and ensure secure, compliant operations in healthcare, manufacturing, and defense contexts.

You will mentor junior engineers and contribute to peer-reviewed research and open-source releases while building user-facing interfaces and secure inference frameworks in regulated

Qualifications

• Master’s degree or higher in Computer Engineering, Computer Science, or Data Science and two (2) years of relevant experience.

Responsibilities

• Architect and build end-to-end agentic AI systems for secure environments.
• Design multi-agent orchestration coordinating distributed agents over a bidirectional transport.
• Implement delegated-authority with auditable actor claims (OAuth 2.0).
• Develop standardized agent communication protocols (MCP, A2A).
• Build a server-driven UI and translation layer for diverse devices.
• Ensure security, compliance, and auditability in high-stakes domains.
• Provide technical leadership and mentoring, and contribute to research/Open-source.
• Operate in environments handling sensitive data with cybersecurity best practices.

Skills

Distributed systems

APIs

Real-time transports

Authentication/Authorization

MCP

A2A

OAuth 2.0 Token Exchange

Mentoring

Publications

Education

Master's degree or higher in Computer Engineering, Computer Science, or Data Science

Job description

University of Louisville seeks a Senior AI Engineer to lead the design and delivery of secure, end-to-end multi-agent AI systems. You will architect orchestration across distributed agents with auditable delegation, and ensure secure, compliant operations in healthcare, manufacturing, and defense contexts.

You will mentor junior engineers and contribute to peer-reviewed research and open-source releases while building user-facing interfaces and secure inference frameworks in regulated