The Product Podcast

Toast was built long before generative AI existed. So how do you weave AI agents into a platform that thousands of restaurants already run their business on, without breaking the thing that works? In this episode of The Product Podcast, Carlos (CEO at Product School) sits down with Maggie Crowley, VP of Product at Toast, to unpack what it actually takes to build AI into mature, pre-AI SaaS.

Maggie is a former Olympic speed skater (fifth at the 2006 Winter Olympics) turned product leader, with an MBA from Harvard and product roles across Drift, TripAdvisor, and Charlie Health before Toast. She was also a Product School contributor years ago, back in her Drift days, which is what makes this a fitting return. At Toast she leads product on Toast IQ, the company's AI layer for restaurant and retail operators. In this conversation, she and Carlos get into the genuinely hard parts: why the quality bar for building on an established platform is punishingly high, how her team actually shipped their first assistant (sitting in WhatsApp groups with real customers and turning it on for a few of them), and why you have to physically bring the AI to the user instead of assuming they'll discover it.

Maggie is refreshingly blunt that restaurant operators have zero tolerance for "AI for AI's sake." They don't care whether it's Anthropic, OpenAI, or an open-weight model under the hood; they only care whether you helped them run their business. She explains how the tight P&L of a restaurant forces real discipline about what's worth charging for, how Toast IQ Grow (an agent that drives demand and fills seats) became a breakthrough use case by slashing the cost of marketing work, and why she still manages a frontline team directly to stay close to how AI is changing the way products get built. The episode closes with a live demo of Toast IQ.

What you'll learn:
- How to build AI agents into a platform designed years before AI, without lowering the quality bar
- The scrappy build process: WhatsApp groups with customers and turning an assistant on for a few
- Why you have to bring the AI to the user, and why AI onboarding is an unsolved problem
- Why context (where the user is, what time, what role) is everything for an AI product
- Why restaurant operators have zero tolerance for "AI for AI's sake"
- How the tight P&L of a restaurant enforces discipline about what's worth building and charging for
- How Toast IQ Grow uses AI to cut the cost of marketing and drive real demand
- Why abstracting model choice (Anthropic vs. OpenAI vs. open weights) away from the user matters
- Why a product leader should keep a hand in how teams build, and manage close to the work
- A live look at Toast IQ answering real operator questions in plain language

Connect with Maggie Crowley:
VP of Product, Toast
LinkedIn: https://www.linkedin.com/in/maggie-crowley-42a97112/

Host: Carlos, CEO at Product School
LinkedIn: https://www.linkedin.com/in/villaumbrosia/

About Toast: Toast is an all-in-one platform built for restaurants and retail, spanning point of sale, payments, operations, and marketing, now including Toast IQ, its AI layer for operators.

About the Product Podcast: Product School's podcast brings you candid conversations with the founders and product leaders shaping tech.
Social Links:
Find out more about Product School here
Follow our Podcast on TikTok here
Follow Product School on LinkedIn here

Most apps optimize for time on screen. Hinge optimizes for you leaving. In this episode of The Product Podcast, Carlos (CEO at Product School) sits down with Ben Celebicic, Chief Product and Technology Officer at Hinge, the dating app "designed to be deleted," to unpack how you build a product whose North Star is great dates, not engagement, and still chase a billion dollars in revenue.

Ben has been at Hinge for 11 years, since the 2015 relaunch, and recently became one of a growing number of leaders to hold both the product and technology mandate in a single CPTO role. He and Carlos get into the healthy tension between the CPO instinct (ship fast, innovate) and the CTO instinct (protect quality and infrastructure), and how he now runs that debate inside one person and across a team of nearly 300. He lays out Hinge's monetization philosophy in two principles: build the free experience so everyone can succeed, and only charge for things you can't give away free without breaking the ecosystem (which is why fewer than 15% of users pay). He explains how a dedicated culture and consumer insights team studies real daters to find core needs, how generative AI now enables per-user profile coaching that was impossible before, why trust and safety makes up roughly a third of the entire Hinge organization, and where he draws the line on AI-assisted engineering today. He closes on Hinge's mission to reach $1 billion in revenue in 2027.

What you'll learn:
- Why "designed to be deleted" is a real product principle, and how an anti-engagement North Star works
- How to run the CPO-vs-CTO tension, even when both roles live in one person
- Hinge's two-principle monetization model, and why fewer than 15% of users pay by design
- Why you should only charge for things that would break the ecosystem if they were free
- How a culture and consumer insights team turns real dater problems into features
- How generative AI enables personalized profile coaching that wasn't possible before
- Why trust and safety is about a third of the whole org, and how AI powers it
- Where an engineering-background leader draws the line on AI-built product today
- The mission to hit $1B in revenue in 2027, and the "obsess over user needs" formula behind it

Connect with Ben Celebicic: CPTO, Hinge
LinkedIn: https://www.linkedin.com/company/lrb-podcasts/about/

Host: Carlos, CEO at Product School
LinkedIn: https://www.linkedin.com/in/villaumbrosia/

About Hinge: Hinge is the dating app "designed to be deleted," built to help people get off the app and out on great dates. Founded by Justin McLeod and relaunched in 2015, Hinge is part of Match Group's portfolio of dating and social discovery apps.

About the Product Podcast: Product School's podcast brings you candid conversations with the founders and product leaders shaping tech.
Social Links:
Find out more about Product School here
Follow our Podcast on TikTok here
Follow Product School on LinkedIn here

One month before Bolt exploded, Eric Simons was preparing to wind the company down. In this episode of The Product Podcast, Carlos (CEO at Product School) sits down with Eric Simons, CEO of Bolt, to tell the story behind one of the fastest revenue ramps software has ever seen, and what he learned building toward it for seven years first.

Eric is a high-school dropout who approaches life from first principles and treats endurance sports as training for the pain of building a company. He and Carlos trace the near-death moment (a board meeting set to start spinning the company down was one month away), the bet on WebContainers (an operating system Eric's team wrote to run in WebAssembly inside the browser, so you can build real full-stack apps from a browser tab, the way Figma brought design to the web), and the ramp that followed: zero to about $5M ARR in the first 30 days, then to roughly $20M ARR within two months, a curve that normally takes companies years. He explains why StackBlitz is now going deep on professional builders (PMs, designers, engineers), how Bolt is solving the hard part of turning AI-generated prototypes into production code your engineering team will actually sign off on, and why he frames the market like the ChatGPT-vs-Anthropic split: pick a specific ICP and workflow instead of spraying money at everything. He closes on the open-weights bet (StackBlitz was the earliest backer of Vite), aggressive usage-based pricing, and why cheaper open models are closing the last bit of distance to the labs.

What you'll learn:
- The near-death story: one month from shutting down, then a record revenue ramp
- Why the WebContainer bet (a real OS running in the browser) took seven years and made Bolt possible
- What a 0-to-$20M-ARR-in-two-months curve actually looks like, and why it's so rare
- Why StackBlitz is focusing on professional product builders, not just cute prototypes
- How to turn an AI prototype into production code your engineering team signs off on
- The "ChatGPT vs. Anthropic" framing: why going deep on one ICP beats spraying money everywhere
- Why first-principles thinking and endurance sport shape how Eric builds
- The open-weights bet, aggressive usage-based pricing, and open models catching up to the labs

Connect with Eric Simons, CEO, Bolt
LinkedIn: https://www.linkedin.com/in/eric-simons-a464a664/

Host: Carlos, CEO at Product School
LinkedIn: https://www.linkedin.com/in/villaumbrosia/

About Bolt / StackBlitz: StackBlitz is the company behind Bolt (bolt.new), an AI-powered web development platform that lets anyone build real full-stack applications from a browser tab, powered by its in-browser WebContainer operating system. StackBlitz was also the earliest backer of the open-source Vite project.

About the Product Podcast: Product School's podcast brings you candid conversations with the founders and product leaders shaping tech.
Social Links:
Find out more about Product School here
Follow our Podcast on TikTok here
Follow Product School on LinkedIn here

What do you do after you've founded a company, sold it to Databricks, and could coast? Amandeep Khurana went back to being hands-on. In this episode of The Product Podcast, Carlos Gonzalez de Villaumbrosia (CEO at Product School) talks with Amandeep Khurana, now on Anthropic's go-to-market team, about a career built on deliberately choosing the harder path.

Amandeep traces the whole arc: training as an engineer, moving to the Valley to go customer-facing at Cloudera, catching the company-building bug "by osmosis," and founding Okera in 2016, which eventually exited to Databricks. He talks about what he learned in the ups and downs of running a startup, why he later joined AWS as a founding PM (on Kiro) to run four zero-to-one initiatives instead of taking a comfortable management track, and how he thinks about the two fundamental jobs in any company: building a product and selling a product. He and Carlos dig into the tension between being a hands-on builder and being a manager, how to decide which domains are worth jumping into when you know nothing, and why he joined Anthropic to work on bringing AI safely to enterprises. It's a candid conversation about career design, first-principles decision-making, and staying close to the work.

What you'll learn:
- Why Amandeep repeatedly chooses the "harder path," and how first-principles thinking drives his career decisions
- What founding and exiting Okera (to Databricks) actually taught him about company building
- Why he went back to hands-on building instead of a pure management track
- How to think about the two core jobs in any company: building the product and selling it
- How to decide whether to jump into a domain you know nothing about
- What running four zero-to-one initiatives inside a big company teaches you
- Why he moved into GTM and enterprise AI adoption at Anthropic
- How to stay hands-on as your career pulls you toward management

Connect with Amandeep Khurana:
Anthropic (GTM / Enterprise); previously co-founder and CEO of Okera, founding PM for Kiro at AWS
LinkedIn: https://www.linkedin.com/in/amansk/

Host: Carlos Gonzalez de Villaumbrosia, CEO at Product School
LinkedIn: https://www.linkedin.com/in/villaumbrosia/

About the Product Podcast: Product School's podcast brings you candid conversations with the founders and product leaders shaping tech.
Social Links:
Find out more about Product School here
Follow our Podcast on TikTok here
Follow Product School on LinkedIn here

Visual collaboration is missing from the AI era. That's the provocation Elaina O'Mahoney, Chief Product Officer at Mural, brings to this episode of The Product Podcast with Carlos (CEO at Product School). We spend all day alone in our tools, typing or talking to an LLM, and when it's time to collaborate we drop words into a doc instead of thinking together visually. Elaina makes the case that AI has quietly turned work into "solo-player mode," and that the teams who win next will be the ones who rebuild shared context and multiplayer collaboration on top of it.

Elaina and Carlos get into why ideas are now cheap (you can go from a thought to a working prototype in hours) and why that makes shared context, not idea generation, the real bottleneck. She argues that everyone is a designer at heart, that product leaders should replace the five-page strategy doc with a fast prototype that conveys the North Star, and that "shipping" can mean a workflow or an agent, not just a production release. They also dig into the crowded "where work happens" landscape (Slack, Teams, Jira, Linear, Notion) and the blank-canvas gap around how and why decisions actually get made, how to create safe spaces for people to try AI and fail inside large enterprises, and the pricing tension between legacy seat-based SaaS and AI-native usage models.

What you'll learn:
- Why Elaina says visual collaboration is "absent" in the AI era, and why that matters
- The "solo-player mode" problem: how AI made work lonely, and what multiplayer should look like
- Why ideas are now cheap, and how that shifts the bottleneck to shared context and faster decisions
- Why the real blank space isn't planning or chat, but how and why decisions get made
- The case that everyone is a designer, and how visual thinking speeds up decisions
- How a CPO ships to demo a direction (a prototype instead of a strategy doc)
- What's still scarce in an AI world: data science, taste, and synthesizing analytics
- How to create safe, low-risk spaces for teams to try AI and fail inside big enterprises
- Why smaller "champion" teams beat top-down rollouts, and how internal builders spread practices
- The pricing dilemma: seat-based SaaS vs. AI-native usage models

Chapters:
00:00 Intro
00:59 Intro: Elaina O'Mahoney, CPO at Mural
01:31 Mural's origin as a collaboration canvas
03:01 Multiplayer vs. AI's solo-player mode
04:23 Ideas are cheap now: idea to prototype in hours
05:44 The competitive landscape and convergence
08:31 The blank space nobody owns: how decisions get made
09:22 Lowering the bar so everyone can design
11:33 Everybody should be shipping now
12:25 Shipping to demo, not to production
14:04 Creating safe spaces to try AI in the enterprise
17:10 Forward-deployed engineers
19:14 Building a community around Mural
22:11 The pricing dilemma: SaaS vs. AI-native usage
24:07 Measuring real AI impact
26:19 How Elaina structures her team
28:20 The innovator's dilemma
30:03 From annual QBRs to monthly loops
31:05 Ten years of Mural, and a World Cup sign-off

Connect with the Guest:
Elaina O'Mahoney, Chief Product Officer, Mural
LinkedIn: https://www.linkedin.com/in/eomahoney/

Host: Carlos, CEO at Product School
LinkedIn: https://www.linkedin.com/in/villaumbrosia/

About Mural: Mural is a visual collaboration platform built for teams to brainstorm, converge, and make decisions together on a shared canvas. Originally created by co-founders in Argentina, it scaled rapidly during the shift to remote work.

About the Product Podcast: Product School's podcast brings you candid conversations with the founders and product leaders shaping tech.

Social Links:
Find out more about Product School here
Follow our Podcast on TikTok here
Follow Product School on LinkedIn here