Product Manager, AI Revenue Systems

Ramp Ramp · Fintech · New York, NY · Product

Product Manager for Ramp's AI Revenue Systems, focusing on building the AI platform that GTM runs on. This involves owning agentic tooling and workflows for GTM teams, the GTM data platform (scoring, routing, data acquisition), and ensuring data accuracy for AI tools. The role emphasizes instrumenting, iterating, and closing the loop with eval frameworks and feedback systems. Requires hands-on AI experience, understanding of LLM concepts, SQL proficiency, and comfort with ambiguity.

What you'd actually do

  1. Build for GTM Teams. Own the agentic tooling and workflows for the GTM teams that are less served today — SDR prospecting and outreach, Solutions discovery and POV workflows, and channel partner execution. These motions are high-context and high-value, and largely greenfield.
  2. Own the GTM data platform. Define the data contracts, schemas, and pipelines that make GTM intelligence possible. Own account scoring, routing, and assignment — the systems that determine which accounts get attention, from whom, and when. Drive net new data acquisitions that expand what our agents can reason over. Ensure the data feeding our AI tools is accurate and trustworthy enough to act on.
  3. Serve every GTM motion, not just one. Resist the pull toward optimizing for a single team. Understand the distinct workflows and incentives across the GTM org and build systems flexible enough to serve all of them — while still being opinionated about what good looks like.
  4. Instrument, iterate, and close the loop. Define what good looks like. Build eval frameworks, feedback systems, and dashboards that tell you whether the tools are driving real adoption and impact — and use that signal to make reps active participants in improving the system over time.

Skills

Required

  • 1–3 years of product experience, or 3 - 5 years of total experience in a role that built real judgment — banking, consulting, deployment strategy, agent PM, GTM operator, or founder.
  • Deep hands-on experience building with AI: you've prototyped, shipped, and iterated on AI tools or agents — not just managed roadmaps about them.
  • Technical fluency with modern AI coding harnesses (Cursor, Claude Code, Codex).
  • Working knowledge of core LLM concepts (prompting, embeddings, retrieval, evals) and the judgment to translate these into reliable products where hallucinations have real consequences.
  • Comfortable in SQL and confident reading data. You can pull your own analysis, spot what the numbers aren't telling you, and make decisions without waiting for a data team.
  • Curious systems thinker who learns fast and defaults to building.
  • Comfort with 0-to-1 ambiguity.

Nice to have

  • Broad GTM fluency and operational empathy for the field
  • Strong working knowledge of GTM data and systems: CRM data models, sales engagement tooling, pipeline data, and intent signals.
  • Prior work in fintech, enterprise SaaS, or other domains where data quality is load-bearing.
  • Proven ability to manage senior stakeholders
  • Curiosity about externalizing internal AI work

What the JD emphasized

  • AI platform that GTM runs on
  • agents handle execution
  • reliable background execution at scale
  • human-agent collaboration that earns trust
  • feedback loops that turn raw GTM signal into a compounding organizational advantage
  • agentic tooling and workflows
  • data contracts, schemas, and pipelines
  • account scoring, routing, and assignment
  • data acquisitions that expand what our agents can reason over
  • data feeding our AI tools is accurate and trustworthy enough to act on
  • build systems flexible enough to serve all of them
  • Build eval frameworks, feedback systems, and dashboards
  • Deep hands-on experience building with AI: you've prototyped, shipped, and iterated on AI tools or agents — not just managed roadmaps about them.
  • Technical fluency with modern AI coding harnesses (Cursor, Claude Code, Codex).
  • Working knowledge of core LLM concepts (prompting, embeddings, retrieval, evals) and the judgment to translate these into reliable products where hallucinations have real consequences.

Other signals

  • AI platform for GTM
  • agents handle execution
  • playbooks encode institutional knowledge
  • feedback loops make system smarter
  • reliable background execution at scale
  • human-agent collaboration
  • scoring and routing systems
  • data infrastructure and acquisitions
  • agents for SDRs, Solutions, and channel teams
  • AI tools
  • LLM concepts
  • hallucinations have real consequences