Skip to main content
Product School Podcast

Zapier VP of Product on Orchestrating 800+ AI Agents to Manage Everything | Chris Geoghegan | E286

33 min episode · 2 min read
·
Chris Geoghegan

Episode

33 min

Read time

2 min

Topics

Leadership, Sales & Revenue, Artificial Intelligence

AI-Generated Summary

Key Takeaways

  • Agent Orchestration at Scale: Zapier operates 800 internal AI agents that trigger autonomously on schedules or real-world events, complete tasks, and report back without human initiation. To build this, treat each agent like a new hire: write a clear job description as the prompt, onboard with relevant context, and assign specific tools via MCP or Zapier actions.
  • API vs. MCP for Agent Tooling: APIs require agents to write recall logic from scratch each time. MCP provides a standardized layer where tools come pre-described with defined inputs and usage instructions, reducing errors and enabling agents to select and execute the right tool reliably across complex, multi-step workflows without manual intervention each session.
  • Competitive Moat Through Use-Case Data: When LLMs release competing automation features, Zapier's defensible advantage is its dataset of real automation use cases across 3.4 million businesses. The strategic play is surfacing that data to answer the enterprise bottleneck question: what should we automate next, rather than competing on raw agent-building capability alone.
  • AI Governance Framework for Enterprise: IT and security teams evaluate AI tools on three criteria: who is sending data, what data is being sent, and where it is going. Product teams selling into enterprise should build observability dashboards and access controls that answer these three questions explicitly, framing them as AI governance rather than a compliance checkbox.
  • Adoption vs. Transformation Distinction: Adoption means using AI to execute existing processes faster. Transformation means doing something previously impossible or unfundable. When measuring enterprise AI value, track both separately. Start clients on adoption metrics like time saved per workflow, then build toward transformation metrics tied to net-new capabilities the organization could not execute before AI.

What It Covers

Chris Geoghegan, VP of Product at Zapier, explains how the company runs 800 internal AI agents across product, engineering, and leadership functions, and outlines how product teams should rethink workflows, enterprise AI governance, and competitive positioning as orchestration replaces traditional automation.

Key Questions Answered

  • Agent Orchestration at Scale: Zapier operates 800 internal AI agents that trigger autonomously on schedules or real-world events, complete tasks, and report back without human initiation. To build this, treat each agent like a new hire: write a clear job description as the prompt, onboard with relevant context, and assign specific tools via MCP or Zapier actions.
  • API vs. MCP for Agent Tooling: APIs require agents to write recall logic from scratch each time. MCP provides a standardized layer where tools come pre-described with defined inputs and usage instructions, reducing errors and enabling agents to select and execute the right tool reliably across complex, multi-step workflows without manual intervention each session.
  • Competitive Moat Through Use-Case Data: When LLMs release competing automation features, Zapier's defensible advantage is its dataset of real automation use cases across 3.4 million businesses. The strategic play is surfacing that data to answer the enterprise bottleneck question: what should we automate next, rather than competing on raw agent-building capability alone.
  • AI Governance Framework for Enterprise: IT and security teams evaluate AI tools on three criteria: who is sending data, what data is being sent, and where it is going. Product teams selling into enterprise should build observability dashboards and access controls that answer these three questions explicitly, framing them as AI governance rather than a compliance checkbox.
  • Adoption vs. Transformation Distinction: Adoption means using AI to execute existing processes faster. Transformation means doing something previously impossible or unfundable. When measuring enterprise AI value, track both separately. Start clients on adoption metrics like time saved per workflow, then build toward transformation metrics tied to net-new capabilities the organization could not execute before AI.

Notable Moment

Geoghegan describes Zapier's executive team spending 90 minutes at a recent leadership offsite doing AI show-and-tell, where each executive demonstrated their personal AI usage live. He argues that leaders who mandate AI transformation without hands-on tool use lose credibility with their teams almost immediately.

Know someone who'd find this useful?

Episode Transcript

What does it mean to do that collectively across all of your agents that you're hiring? So we have a new workforce, essentially, of agents who are doing work on our behalf that we're hiring to do this job. Orchestration is how do all those pieces work together? We have about 800 of these agents at Zapier that are doing work on our behalf, triggering on different schedules or events that are happening in the real world, going and doing the work and coming back and say, hey, I did this work. What do you think? One, the very first thing we did was just call the code red. I mentioned that a little bit earlier. But I think it's important to do something that says, hey, we're in a moment and this moment is not like any other moment and we need to do something about it. Who is doing it? What is the data? And where are they sending it? Do we have the right policies? Are those policies enforced? And then can I see that those policies are enforced? Hey. This is Carlos, CEO at Product School and your host on the product podcast. Today's guest is Chris Gaugigan, VP of product at Zapier, the automation platform that has become the connective tissue for over $3,400,000 businesses ranging from early stage startups to 69% of the Fortune 1,000. Zapier allows anyone to connect over 7,000 different applications to to automate complex workflows without writing a single line of code. They have scaled into a $5,000,000,000 valuation with over $400,000,000 in ARR, all while remaining profitable since 2014. During our conversation, we explore how they are orchestrating over 800 AI agents internally to manage everything from complex product demos to engineering triage, giving their team unprecedented leverage, how product managers can shift their mindset from designing rigid UI flows to probabilistic AI outcomes, how to build a moat to not die when LLMs like ChatGPT or Claude release a competitive feature. Let's dive in. Welcome to the product podcast, Chris. Awesome. Thanks for having me, Carlos. Chris, as a fun fact, I've been using Zapier for many, many years way before AI was cool. And it was funny when I realized that you've been there for almost ten years, and you were the first ever PM there. Right? Yeah. Yeah. It's, it's been quite a journey. I joined obviously, the founders were leading the product, and I joined, Mike Canoop, one of the founders hired me and said, hey. What what's the first feature you wanna build? And so the first the very first thing I worked on as a PM at Zapier was Zapier for Teams. Up until that point, it had just been like a solo player, product. So that was my first project. It's always so hard for a founder, especially like a product minded founder, to hire a PM and kind of delegate the baby a little bit. So how were you able to …

Get the full transcript (6,319 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all Product School Podcast transcripts →

You just read a 3-minute summary of a 30-minute episode.

Get Product School Podcast summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links.

Tools

  • by Zapier

    Chris Geoghegan, VP of Product at Zapier, explains how the company runs 800 internal AI agents across product, engineering, and leadership functions
  • MCP provides a standardized layer where tools come pre-described with defined inputs and usage instructions, reducing errors and enabling agents to select and execute the right tool reliably

More from Product School Podcast

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best Product Management Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's AI & Machine Learning Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into Product School Podcast.

Every Monday, we deliver AI summaries of the latest episodes from Product School Podcast and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime