Product Manager, Agentic Tools & Orchestration
About The Position
ZoomInfo is where careers accelerate. We move fast, think boldly, and empower you to do the best work of your life. You'll be surrounded by teammates who care deeply, challenge each other, and celebrate wins. With tools that amplify your impact and a culture that backs your ambition, you won't just contribute. You'll make things happen—fast. ZoomInfo isn't adding AI to our products—we've rebuilt around it. Our multi-agent orchestration platform powers intelligent automation across every product surface: Copilot, GTM Studio, MarketingOS, and beyond. Now we need someone to own the capability layer that determines what our agents can actually do. This role owns the agentic tools ecosystem—the MCP servers, LangChain integrations, and composable building blocks that give our agents their superpowers. Every new capability our agents gain flows through your platform. Every integration that connects ZoomInfo's intelligence to the broader ecosystem ships through your roadmap. When our agents book meetings, enrich records, trigger sequences, or orchestrate complex GTM workflows, they're executing tools you've defined, optimized, and deployed. The agentic tools landscape evolves weekly. New standards emerge. New integration patterns prove themselves. New orchestration primitives unlock capabilities that seemed impossible months ago. You'll operate at this frontier—evaluating what's production-ready versus what's research curiosity, and making the calls that shape how ZoomInfo's agents interact with the world. This is infrastructure work with exponential leverage. Build one tool well, and every agent across every product benefits. Define one integration pattern correctly, and it becomes the template for dozens that follow. Your architectural decisions don't just ship features—they compound.
Requirements
- 6-15 years of product management experience with 2+ years focused on AI/ML infrastructure, developer platforms, or integration products
- Direct experience with production agentic systems—you've shipped products where LLMs invoke tools, not just generate text
- Hands-on familiarity with agentic frameworks (LangChain, LangGraph, CrewAI, or similar) and emerging standards (MCP, function calling patterns)
- Track record building platform products where engineering teams are your customers
- Experience with integration architecture—APIs, webhooks, authentication patterns, rate limiting, error handling at scale
- History of defining product direction in fast-moving technical domains where you're often months ahead of established best practices
- Applied AI Practitioner You've moved beyond AI hype into practical application. You understand why agents fail—context window limits, hallucinated tool calls, poor error recovery—and you've built systems that work anyway. You can discuss prompt engineering trade-offs, evaluate tool description effectiveness, and debug agent execution traces. Python proficiency lets you prototype, validate assumptions, and speak credibly with engineers.
- Context Engineering Intuition Tools are only as good as the context agents have when invoking them. You understand how to design tool interfaces that give agents what they need to succeed—clear descriptions, appropriate parameters, useful error messages. You think about context budgets, information density, and the cognitive load you're placing on the orchestrating model.
- Integration Systems Thinker You've built products that connect systems. You understand the unglamorous realities: authentication flows, rate limits, webhook reliability, schema evolution, partial failures. You know that "integrate with Salesforce" is a six-month roadmap item, not a two-week task. You design abstractions that hide complexity without sacrificing capability.
- Platform Product Mindset Your customers are other engineers and product teams. You measure success through their velocity—how quickly they can ship new agent capabilities using your platform. You understand platform economics: each team adopting your tools increases their value through shared learnings, amortized maintenance, and network effects. You resist one-off solutions that don't generalize.
- Intellectual Velocity You read agent papers, follow framework releases, and prototype new patterns before they're mainstream. You maintain strong opinions weakly held—your architectural assumptions update as evidence accumulates. The agentic landscape moves too fast for static expertise; continuous learning is non-negotiable.
- Strategic Communication You translate between technical depth and business impact fluently. You can explain to executives why adopting a new tool standard takes three months but prevents six months of technical debt. You can communicate to engineers why business constraints require shipping "works reliably" before "works optimally." You influence through clarity, not authority.
Responsibilities
- Own the Agentic Tools Platform Define, build, and evolve the tool ecosystem that powers ZoomInfo's multi-agent orchestration system. You'll own the registry of capabilities agents can invoke, the standards for how tools are built, and the governance for how they're deployed. Balance the tension between enabling rapid tool development and maintaining the reliability that production agents demand.
- Architect Integration Strategy ZoomInfo's agents don't operate in isolation—they connect to CRMs, marketing automation platforms, communication tools, and data providers. Own the integration roadmap that expands agent capabilities through MCP servers, API connections, and partner ecosystems. Determine which integrations unlock disproportionate value, negotiate technical partnerships, and define the abstraction layers that make integrations composable rather than brittle.
- Drive Agentic Execution Excellence Tools are only valuable if agents use them effectively. Partner with ML engineers on tool selection optimization, prompt engineering for reliable tool invocation, and error handling patterns that make agents resilient. Own the feedback loops that surface tool failures, identify capability gaps, and translate production signals into platform improvements.
- Enable Internal Tool Development Multiple product teams need to extend agent capabilities for their specific domains. Build the developer experience—documentation, templates, testing frameworks, deployment pipelines—that lets teams ship new tools without deep platform expertise. Measure success through tool development velocity: how quickly can a new capability go from idea to production agent invocation?
- Navigate the Agentic Frontier The standards are still being written. MCP evolves. New orchestration patterns emerge from research. Tool-use benchmarks reveal new optimization opportunities. You'll continuously evaluate innovations—function calling improvements, multi-tool planning, tool composition patterns—determining which advances warrant immediate adoption versus patient observation. Bring external best practices to ZoomInfo while contributing learnings back to the community.
- Orchestrate Cross-Functional Execution Translate between worlds: ML engineers optimizing agent reasoning, platform engineers building scalable execution infrastructure, and product teams shipping customer-facing automation. Establish prioritization frameworks that balance "keep the lights on" reliability work against strategic capability expansion. Influence without authority through data, clear reasoning, and earned technical credibility.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Mid Level
Education Level
No Education Listed
Number of Employees
501-1,000 employees
