Agent Product Manager (Remote - US East Coast)

About the role

At Quadrivia, we're pioneering a new way to run a company by freeing brilliant minds to focus on what matters most—patient care—while AI agents handle the repetitive, mindless processes that currently consume clinicians' time. As an Agent Product Manager, you will own the agents deployed in real health system accounts, working on the ground with customers to ensure these agents perform effectively.

Every Quadrivia customer runs on agents—whether it's a discharge follow-up agent for a cardiology service line, an ED intake agent for a rural hospital, or an appointment-reminder agent for a clinic group. This role owns one or more account's agents end-to-end: configuring, testing, deploying, monitoring, and improving them. You will be on-site with customers, wiring agents into their real environments, and staying accountable for their performance. This is a forward-deployed role combined with product ownership, ensuring accountability isn’t lost in handoffs.

You will carry live health system accounts in the U.S. and other markets, owning the automation and its performance. While you won’t manage the implementation program or commercial relationship, you will ensure the agents deliver the outcomes they were designed for. Our tooling is evolving rapidly, and your work will directly shape the templates, test suites, and tooling requirements that make agent-building easier over time. The goal is to make agent management self-serve for customers, and your role is critical to achieving that.

Responsibilities

• Deploy:
  • Go on-site with customers, observe how work is done (e.g., schedulers, care coordinators, practice managers), and translate those workflows into agent behavior. Expect regular U.S. travel.
  • Wire agents into the customer’s real environment, including EHR data/workflows, phone systems, scheduling platforms, insurance/eligibility steps, and clinical protocols. Collaborate with integration engineers to ensure agents have what they need.
  • Serve as the AI and technical voice in customer deployments, sales demos, proofs of concept, and scoping conversations.
• Build:
  • Configure agents in Cortex (our agent platform) end-to-end: conversation graphs (nodes, edges, transition conditions), prompts, variable extraction, escalation, and confirm-then-act logic.
  • Build and maintain journeys—the orchestration around agents, including scheduling, retries, channel selection, call windows, and escalation paths.
  • Define core KPIs and metrics that align with improving customer outcomes and goals.
  • Collaborate with the broader product team to improve the platform and localize tools for new markets/segments.
• Prove:
  • Create durable regression suites in Starfish (our testing toolkit) covering real agent behavior, not just the happy path. Include scenarios, personas, judges, and campaigns.
  • Reproduce production failures as exact scenarios, A/B test fixes, and regression-gate the entire suite before shipping.
  • Work with Clinical and Quality teams to ensure "passing" tests map to clinical safety. Clinical review is a gate for every behavior change—no one approves their own prompt changes.
• Own the outcome:
  • Own the agents on your accounts for as long as they are live. Investigate and fix issues, ensuring behavior is correct, not just explained.
  • Own the metrics that determine automation success (e.g., containment, resolution, escalation rate, clinical quality flags) and report them to customers.
  • Distinguish between agent misconfigurations and platform bugs, escalating the latter to Engineering with evidence.
• Scale:
  • Manage roughly 20 agents across your accounts, building patterns and instantiating them rather than hand-crafting each one.
  • Turn repeated manual work into tooling requirements for Product and Engineering—you are their most important internal customer.
  • Write templates, standards, and definitions of done for agent quality that the rest of the company builds against.

Requirements

• US healthcare fluency, from the inside: You have worked with or inside U.S. provider organizations and understand how health systems operate—scheduling, referrals, front desk/call center workflows, EHR systems, eligibility, prior authorization, and why clinical staff trust or distrust new systems. You know the difference between what a health system says it wants and how work actually gets done.
• Location and travel: US-based on the East Coast (Eastern timezone) and able to travel to customer sites in the U.S. regularly. Pacific-timezone candidates are not a fit due to overlap needs with customers and our European team.
• Experience: 3–6 years in a role demanding fast mastery of an unfamiliar technical domain in front of customers, such as forward-deployed/solutions engineering, technical product management, top-tier consulting, or an early-stage company where you figured things out and shipped them.
• LLM fluency: Demonstrated experience with LLM systems in practice—prompt design, structured extraction, evaluation, and understanding why agents fail. You’ve shipped something and formed opinions based on that experience.
• Technical comfort: Ability to read logs, dashboards, traces, and configuration files, and reason about systems from telemetry. SQL and scripting are expected; writing production code is not required but welcome.
• Communication: Exceptional written communication. Most of your output will be specifications, test cases, root-cause explanations, or customer-facing reports that others must act on.
• Clinical rigor: You take seriously the responsibility of changing systems that interact with patients. While you don’t need a clinical background (we have clinicians), you must prioritize correctness in a clinical context.

Preferred Qualifications

• Hands-on experience with Epic, Cerner/Oracle Health, athenahealth, or similar EHR systems, and with contact center telephony.
• Experience with enterprise customers and stakeholder management.
• Conversational AI or voice-stack experience (ASR/TTS, LiveKit, Twilio, or equivalent).
• Experience building evaluation or regression-testing systems for non-deterministic software.
• A second working language, particularly Spanish or Mandarin.

How we’ll evaluate you

One stage of the process is a build exercise. You’ll get access to Cortex—our real platform, not a sandbox—along with a use-case brief and documentation. Your task is to build an agent and the journey around it: conversation graph, extraction/escalation logic, and orchestration (e.g., when it calls, retries, or hands off to a human). We’ll observe how far you get, what questions you ask, and how you handle gaps in documentation. We care more about your approach to unfamiliar challenges than what you already know. This exercise is also a preview of the job itself, as this is the core of what you’ll do.

Why Quadrivia

• Mission-driven work at the intersection of AI, healthcare, and enterprise technology, with the potential to make healthcare more accessible and affordable for everyone.
• Own the layer where our product meets patients, at a defining moment for AI in healthcare.
• Competitive compensation, equity, and benefits.