AI Agent Engineer - Remote Eligible

AI Agent Engineer (Remote)

A rapidly scaling leader in the telehealth and AI space is seeking an AI Agent Engineer to join their remote team. We are committed to leveraging technology to improve access to essential services and are looking for a hybrid problem-solver to drive our next phase of AI-driven workflow transformation.

This full-time, remote (US-based) role is critical for expanding agentic workflows that reduce manual effort and improve service accessibility. You will be instrumental in the end-to-end lifecycle of our AI agents, from initial discovery to operational management and continuous improvement.

We are looking for a unique combination of skills:

• The Investigator: Dive deep with stakeholders (end-users and executives) to map existing workflows, identify pain points, and translate messy processes into precise agent tasks, policies, and success metrics.
• The Sprinter: Rapidly develop and deploy agents using a low-code AI agent orchestration environment. You'll manage a roster of 15+ agents, driving their continual improvement and measurable impact.
• The Technician: Traverse internal APIs and self-serve to build and refine internal API action endpoints, safely integrating agents into our technical ecosystem.

Key responsibilities include:

• Discover & Scope: Collaborate with clinicians, support, and billing specialists to define the real job-to-be-done and translate complex workflows into precise agent requirements.
• Design Agent Behavior: Author low-code agent specifications (e.g., JSON examples, context graphs, tool lists) to encode instructions, guardrails, allowed actions, and escalation rules.
• Ship Quickly with Guardrails: Establish rapid pilots with robust, deterministic orchestration (states, retries, timeouts), human-in-the-loop paths, and essential performance monitoring.
• Integrate Safely: Connect agents to internal APIs and approved tools, ensuring least-privilege access in partnership with the engineering team to accelerate time-to-impact.
• Evaluate & Improve: Create lightweight evaluation sets, define measurable outcome metrics, and own the full performance lifecycle from insights gathering to iterative enhancement.
• Operationalize: Quantify the ROI (e.g., hours saved, error-rate reduction) and transition stable agents to internal owners with comprehensive runbooks and on-call procedures.