Lead AGENTIC AI Engineer

Orbis is seeking a manager to lead the Mission Support Desk and owns the integration and operation of agentic AI capabilities across Orbis programs. You are responsible for the team's output, the quality and reliability of AI-assisted workflows, and the continuous expansion of automation that frees analysts and program staff to focus on mission rather than process. You will design and deploy agentic AI workflows, manage a team of support specialists, and serve as the internal authority on what AI can and cannot do well in a support context.

This role is enterprise-wide in scope. Every program that depends on the Mission Support Desk depends on the quality of what you build and the judgment of the team you lead. Roughly 60% of your time is direct delivery and AI workflow development; 40% is people management, planning, and cross-program coordination. This is not a research role and not a purely managerial one.

You are accountable for output that ships and for systems that work.

• Own the desk's delivery. Be accountable for the full output of the Mission Support Desk across all programs it serves - response quality, turnaround time, and the accuracy of AI-generated outputs. The desk's performance is your performance.
• Design and operate agentic AI workflows. Build, refine, and maintain the agentic pipelines that support desk operations - including triage, data retrieval, summarization, reporting, and escalation routing. Own the configuration, prompt architecture, and reliability of every workflow in production.
• Integrate across Orbis products. Work with Catalyst, Pulse, and Discovery to ensure the support desk can access and act on data flowing through the governed data fabric - surfacing the right information at the right time, in the right format, for the right consumer.
• Set the quality bar. Define what "good" looks like for AI-assisted support output - accuracy standards, human review checkpoints, escalation criteria, and documentation requirements. Hold the team and the systems to it.
• Manage and develop the team. Lead a team of support specialists. Identify skill gaps, give direct performance feedback, and build pathways for growth as AI tooling evolves. The people you develop are the capability you leave behind.
• Build and maintain playbooks. Author and continuously improve the standard operating procedures, prompt templates, agent configurations, and escalation runbooks that govern how the desk operates. Nothing should live only in someone's head.
• Track and report performance. Own the metrics that matter - resolution rate, escalation rate, AI output accuracy, response time, and program satisfaction. Report clearly to program leadership and internal stakeholders, with proposed actions, not just status.
• Identify and close capability gaps. Recognize when a support need is not served by existing AI tooling, scope the gap clearly, and work with engineering or product teams to close it. Surface problems early with proposed mitigations.
• Coordinate across programs. Serve as the operational point of contact for program teams consuming support desk services. Translate their requirements into workflow updates and configuration changes without disrupting active operations.
• Stay current. Track developments in agentic AI tooling, LLM capabilities, and enterprise AI platforms. Assess what is worth adopting, at what pace, and for which programs - and make those calls with confidence.

• 5+ years of professional experience in operations, technical program management, managed services, or AI/data roles
• Demonstrated experience designing and operating workflows powered by LLM-based agents or agentic AI frameworks (e.g., Claude, GPT-4, Lang Chain, Auto Gen, or equivalents)
• Proven ability to manage a team and be accountable for its collective output, not just individual contributions
• Strong working knowledge of prompt engineering, retrieval-augmented generation (RAG), and AI agent orchestration patterns
• Experience working with enterprise API ecosystems – REST, GraphQL, or pub/sub architectures – at production scale
• Ability to assess and communicate the reliability and limitations of AI…