Supra Insider

What if the most important skill for building AI products has nothing to do with evals, technical background, or knowing how to write a prompt? What if it is the ability to design systems that can handle what you never planned for?
In this episode of Supra Insider, Marc Baselga and Ben Erez sit down with Apurva Garware, who has built and scaled products across Amazon, Microsoft, and Upwork, to make the case that systems thinking is the defining skill of the next era of product management. Apurva explains why non-determinism forces PMs to stop thinking in features and start designing the guardrails, agent contracts, and escalation points that govern how a system behaves at runtime, when no one is watching.
They explore a three-phase framework for governing AI systems across design, deployment, and production; heuristics for deciding what to hand to agents versus escalate to humans; and a sharp insight about the two products every AI-native company is actually building: the customer-facing product, and the internal operational system that drives margin and velocity. Marc and Ben also share their own experience calibrating an agentic workflow at Supra, grounding the conversation in practice.
If you are a PM trying to find your footing in the AI era without a deeply technical background, a founder wrestling with when to reach for AI versus simpler deterministic automation, or a product leader who wants to build more discipline into how your team ships AI products, this episode is for you.
All episodes of the podcast are also available on Spotify, Apple and YouTube.
New to the pod? Subscribe below to get the next episode in your inbox 👇
Want company-specific interview intel?
If you’re preparing for PM interviews at Meta, Anthropic, OpenAI, Google, Stripe, DoorDash, Uber, or Figma… we think you’re going to love Insider Loops. You can read about it in Ben’s LI post, Marc’s LI post or check out the site directly below.
In this episode, we covered the following topics:
* (02:13) Apurva’s Amazon story: the Beyoncé/Lemonade moment that exposed how a perfectly functioning algorithm failed due to missing broader context
* (03:47) Why linear thinking is not enough for AI products and why non-determinism requires a fundamentally different mindset
* (07:55) Ben’s working definition of systems thinking: never building anything in isolation, thinking in 3D maps, and setting the laws of physics that govern how a system operates
* (10:05) The NASA control room analogy: what it actually means to govern AI agents at scale with confidence
* (11:15) Three phases of AI system governance: design-time contracts, deployment benchmarks, and runtime production escalations
* (17:20) The 49/51 decision rule: a heuristic for knowing which decisions to escalate to humans and which to trust the system to handle
* (21:25) The “two products” insight: every AI-native company is simultaneously building a customer-facing product and an internal operational product, and most are treating the second one as an afterthought
* (26:26) Rule of thumb: if a deterministic automation can do the job, build that instead of reaching for an AI agent
* (32:33) How to start developing systems thinking: automate personal workflows, follow industry thought leaders, and stay curious even when your day job does not involve AI products
* (39:05) The biggest failure state: trying to build the entire system end-to-end from day one, and why scoping is more important than ever
* (41:00) The pre-mortem approach: scoping what good looks like, then mapping every way it could break before you build
* (51:00) Outsourcing thinking versus reasoning: why PMs must keep their own reasoning intact even as AI handles more of the execution
* (56:11) Gratitude corner: mentors, product communities including Supra, and past teams
* And more!
Links:
* Andrew Ng: https://www.linkedin.com/in/andrewyng/
* Dario Amodei: https://www.linkedin.com/in/dario-amodei-3934934/
* Sam Altman: https://x.com/sama?lang=en
* Aaron Levie: https://www.linkedin.com/in/boxaaron/
* Andrej Karpathy: https://www.linkedin.com/in/andrej-karpathy-9a650716/
* Ilya Sutskever: https://www.linkedin.com/in/ilya-sutskever/
* John Hegeman: https://www.linkedin.com/in/john-hegeman-02137b7/
* Supra: https://www.joinsupra.com/
* Products that count: https://productsthatcount.com/
* The Skip: https://theskip.substack.com/
* Apurva Garware: https://www.linkedin.com/in/apurvacg/
* Marc: https://www.linkedin.com/in/marcbaselga/
* Ben: https://www.linkedin.com/in/benerez/
If you’re a new subscriber, we encourage you to check out some of the recent episodes you might have missed:
To support the podcast, please check out the links below:
* Check out Ben’s AI Practice Copilot and top-rated Maven course: Self-Paced PM Interview Bootcamp: Product Sense & Analytical Thinking (selected by Lenny Rachitsky as a top course 🔥). Get 25% off the course with promo code suprainsider (auto-applied here).
* Supra has teamed up with Maven to bring you something special – courses that our own community members have personally curated. And because you’re part of the Supra family, you can get up to 35% in some of these handpicked selections with code SUPRAxMAVEN.


This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit suprainsider.substack.com

What happens to the product development process when the lines between who builds, who designs, and who decides start to disappear?
In this special episode of Supra Insider, recorded as part of the Blurring Lines series with Aster AI, Ben Erez sits down with Rishin Banker, VP of Product at Maven, to explore how a 25-person team is rethinking product development in real time. Rishin opens with a concrete shift: Maven went from two to three concurrent projects to five to six, same headcount, smaller pods, more decision-making at the team level. The unlock wasn’t hiring. It was front-loading strategy so more people could move into the build phase at once.
They explore how Maven’s head of design shipped a full marketing page to production end-to-end, why months of foundational design system work made that possible, and where Figma still fits. Rishin also gets into the tensions he’s navigating, unexpected handoffs, competing priorities when people build in silos, and the difference between projects that can live in their own container versus ones that need specialist input from the start.
If you’re a product leader restructuring your team for the current moment, a designer or PM excited about building more but unsure how to navigate the role blurring, or curious how a lean startup is actually operationalizing these changes day to day, this episode is for you.
A special thanks Alex Pavlou and our friends at Aster AI for hosting this session!
All episodes of the podcast are also available on Spotify, Apple and YouTube.
New to the pod? Subscribe below to get the next episode in your inbox 👇
Want company-specific interview intel?
If you’re preparing for PM interviews at Anthropic, OpenAI, Google, Stripe, DoorDash, Uber, or Figma… we think you’re going to love Insider Loops. You can read about it in Ben’s LI post, Marc’s LI post or check out the site directly below.
In this episode, we covered the following topics:
* (01:05) Intro to the Blurring Lines series and what Maven’s team structure looks like today
* (03:17) From 2-3 to 5-6 concurrent projects: how Maven restructured into smaller pods
* (05:23) Front-loading strategy to unlock parallel building, why the bottleneck definition is shifting
* (07:47) Non-EPD team members becoming builders: the marketing lead vibe-coding a landing page
* (09:24) The designer who shipped end-to-end to production, and why that required foundational work first
* (12:23) Why good engineering hygiene (clean APIs, design systems) matters even more in the AI era
* (17:40) Where Figma still fits, and where clickable prototypes are replacing it
* (21:35) How the team replaced standing meetings with live reviews and more frequent huddles
* (22:53) CEO office hours replacing the weekly design review, and the Duolingo daily review comparison
* (26:00) Change management: what PMs, designers, and engineers are each feeling right now
* (27:50) The real tensions: unexpected handoffs and mismatched ownership between silo builders
* (33:45) How to manage up and bring your CEO along in the current moment
* (34:25) Friday demos: Maven’s favorite meeting of the week and why it’s almost always an AI showcase
* (36:47) Closing thoughts: the silo vs. collaboration tension Rishin is still working through
* And more!
Links:
* Rishin Banker: https://www.linkedin.com/in/rishinbanker/
* Maven: https://maven.com/
* Our episode with Gagan Biyani (CEO of Maven): https://suprainsider.substack.com/p/98-why-mid-career-people-are-doubling?r=486xld
* Aster AI: https://www.asterhq.com/
* Alex Pavlou (Co-founder, Aster AI): https://www.linkedin.com/in/alexpavlou/
* Marc: https://www.linkedin.com/in/marcbaselga/
* Ben: https://www.linkedin.com/in/benerez/
If you’re a new subscriber, we encourage you to check out some of the recent episodes you might have missed:
To support the podcast, please check out the links below:
* Supra has teamed up with Maven to bring you something special – courses that our own community members have personally curated. And because you’re part of the Supra family, you can get up to 35% in some of these handpicked selections with code SUPRAxMAVEN.
* Check out Ben’s AI Practice Copilot and top-rated Maven course: Self-Paced PM Interview Bootcamp: Product Sense & Analytical Thinking (selected by Lenny Rachitsky as a top course 🔥). Get 25% off the course with promo code suprainsider (auto-applied here).


This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit suprainsider.substack.com

What if the reason you feel like you never have enough time isn’t actually a time problem at all?
In this episode of Supra Insider, Marc Baselga and Ben Erez sit down with Molly Siemers, an executive coach for product leaders who spent over two decades building mission-driven products at companies like Kiva, Change.org, and Blurb before launching her coaching practice, Product Craft Works. Molly opens by naming what she's watching in real time: product leaders are running faster than ever, layoffs are everywhere, and the pressure to adopt AI on top of everything else is creating a new kind of cognitive overload. Coaching, she argues, has never been more necessary, not because people need tactics, but because most people are solving the wrong problem.
They explore the difference between time and capacity, why the best senior leaders seem unflappable, and how personal capacity is something you build, not something you find by rearranging your calendar. Molly walks through the integral coaching methodology she trained in, the threshold practice she gives clients to start tapping into body and emotional intelligence, and the three-step framework she keeps returning to: notice, decide, act. The conversation then turns personal, with Marc and Ben reflecting on agency, identity, and what happens when you look around and realize you’ve built a job you hate, or, on the flip side, a life that actually works.
If you’re a product leader feeling overwhelmed and can’t figure out why, someone navigating a career transition and struggling with identity, or a founder or operator who’s curious whether coaching might actually be worth it, this episode is for you.
All episodes of the podcast are also available on Spotify, Apple and YouTube.
New to the pod? Subscribe below to get the next episode in your inbox 👇
Want company-specific interview intel?
If you’re preparing for PM interviews at Anthropic, OpenAI, Google, Stripe, DoorDash, Uber, or Figma… we think you’re going to love Insider Loops. You can read about it in Ben’s LI post, Marc’s LI post or check out the site directly below.
In this episode, we covered the following topics:
* (00:01:34) Why coaching intensity has spiked in the last six weeks: AI acceleration, layoff anxiety, and cognitive overload
* (00:03:17) Who Molly coaches and what makes someone a great coaching client
* (00:05:25) What pulled Molly from product leadership into coaching — and why she always should have been one
* (00:07:04) Does experience matter for coaches? The cold start problem and shared language
* (00:12:54) Coaching fit is about vibes, not credentials — how Molly evaluates both sides of a first conversation
* (00:14:30) “Real time modulation of therapeutic alliance” — Ben’s framework for what makes coaching actually work
* (00:20:36) Assignments as AB tests on your operating system
* (00:22:44) The threshold practice: how to start tapping into emotional intelligence between sessions
* (00:27:56) The three-step coaching framework: notice, decide, act
* (00:37:35) It’s never a time problem — it’s a capacity problem, and why that distinction matters
* (00:40:12) How capacity feels as a concept and how to start building it
* (00:41:23) Agency and capacity go hand in hand — and why Marc does more now than in any full-time role
* (00:57:07) The broken career ladder: what happens when you reach SVP and realize you hate it
* (00:58:50) Why coaching is more important than ever: anchoring to internal signals in a world that changes weekly
* (01:02:15) Where to find Molly and how the audience can be helpful
* (01:03:40) Gratitude Corner: Noah and Mariah DeLeon
* And more!
Links:
* Molly’s Coaching website: https://productcraft.works/
* Molly’s Substack: https://productcraftworks.substack.com/
* Molly Siemers: https://www.linkedin.com/in/mollysiemers/
* Noah Richardson: https://www.linkedin.com/in/noahrichardson/
* Mariah DeLeon: https://www.linkedin.com/in/mariahdeleonhr/
* Island by Aldous Huxley: https://a.co/d/0ek7GmpB
* Marc: https://www.linkedin.com/in/marcbaselga/
* Ben: https://www.linkedin.com/in/benerez/
If you’re a new subscriber, we encourage you to check out some of the recent episodes you might have missed:
To support the podcast, please check out the links below:
* Check out Ben’s AI Practice Copilot and top-rated Maven course: Self-Paced PM Interview Bootcamp: Product Sense & Analytical Thinking (selected by Lenny Rachitsky as a top course 🔥). Get 25% off the course with promo code suprainsider (auto-applied here).
* Supra has teamed up with Maven to bring you something special – courses that our own community members have personally curated. And because you’re part of the Supra family, you can get up to 35% in some of these handpicked selections with code SUPRAxMAVEN.


This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit suprainsider.substack.com

What if you could know whether your product change was going to work — before a single real user ever saw it?
In this episode of Supra Insider, Marc Baselga and Ben Erez sit down with Tom Charman, co-founder of Blok, a synthetic user simulation platform that lets product teams test interfaces, onboarding flows, and product changes against AI-powered behavioral personas, not just to predict conversion, but to model second and third order effects like churn, confusion, and long-term retention. Tom opens by naming what’s changed: shipping is faster than ever, but the tools PMs use to decide what to build haven’t kept up. AB tests still take two to three weeks. Traditional user research still skews toward power users. And as personalization gets more complex, getting to statistical significance gets harder. That’s the gap Blok is trying to close.
They explore how Blok’s behavioral personas go beyond demographics to model psychographics, emotional state, and memory, including what happens when a user has a bad first experience and comes back to your product skeptical. Tom walks through the ICP trap that kills retention, why the “show don’t tell” principle protects against confirmation bias, and why he thinks the biggest shift happening in product right now is the move from reactive to proactive, replacing AB tests and feature flags with pre-ship simulation that gets you to directional confidence in 12 minutes instead of three weeks.
If you’re a PM tired of going to battle in product reviews without data to back your intuition, a founder trying to understand how synthetic users actually work and whether they’re worth it before Series A, or a product leader looking to understand how the entire feedback loop from idea to deployment is about to be rebuilt, this episode is for you.
All episodes of the podcast are also available on Spotify, Apple and YouTube.
New to the pod? Subscribe below to get the next episode in your inbox 👇
Want company-specific interview intel?
If you’re preparing for PM interviews at Anthropic, OpenAI, Google, Stripe, DoorDash, Uber, or Figma… we think you’re going to love Insider Loops. You can read about it in Ben’s LI post, Marc’s LI post or check out the site directly below.
In this episode, we covered the following topics:
* (01:40) Why PMs are becoming the bottleneck: shipping is fast, deciding what to ship isn’t
* (04:00) The two problems with traditional user research: paid participants and sampling bias
* (08:19) Ben’s summary of Blok: how synthetic users work at a high level
* (11:04) Tom’s “show don’t tell” principle and how Blok actually simulates behavior vs. just asking questions
* (13:32) The four layers of a behavioral persona: demographics, psychographics, emotion, and memory
* (16:21) The anxious user example: how emotional state changes conversion behavior step by step
* (20:00) Second and third order effects: why optimizing conversion can quietly kill retention
* (21:52) The ICP trap: why the lower-converting variant might actually be the right call
* (28:05) The agentic loop: how the full product development stack is being rebuilt
* (33:27) Confirmation bias: how Blok’s chat interface helps remove it from the experiment design
* (42:19) Who should be using this: Tom’s honest answer on post-Series A timing
* (46:26) Scrappy alternative vs. Blok: 30–60% accuracy with Claude versus 87% with purpose-built models
* (53:15) How to build trust with skeptical stakeholders: run Blok alongside a real AB test
* (55:41) The biggest shift in product: proactive vs. reactive, and why feature flags may become obsolete
* (58:53) Where to find Tom and how to be helpful to Blok
* (01:00:48) Gratitude Corner: Olivia (co-founder), the team, and investors Hannah, Ashkan, Marlon, Kevin, and Brendan
* And more!
Links:
* Blok: https://www.joinblok.co/
* Tom Charman: https://www.linkedin.com/in/tcharman/
* Olivia Higgs (Tom’s Co-founder): https://www.linkedin.com/in/oliviahiggs/
* Hannah Chelkowski: https://www.linkedin.com/in/hannahchelkowski/
* Ashkan Mizani: https://www.linkedin.com/in/ashkanmizani/
* Marlon Nichols: https://www.linkedin.com/in/marloncnichols/
* Kevin Novak: https://www.linkedin.com/in/novakkm/
* Brendan Baker: https://www.linkedin.com/in/brendanbaker/
* Chris Neumann: https://www.linkedin.com/in/ckneumann/
* Marc: https://www.linkedin.com/in/marcbaselga/
* Ben: https://www.linkedin.com/in/benerez/
If you’re a new subscriber, we encourage you to check out some of the recent episodes you might have missed:
To support the podcast, please check out the links below:
* Supra has teamed up with Maven to bring you something special – courses that our own community members have personally curated. And because you’re part of the Supra family, you can get up to 35% in some of these handpicked selections with code SUPRAxMAVEN.
* Check out Ben’s AI Practice Copilot and top-rated Maven course: Self-Paced PM Interview Bootcamp: Product Sense & Analytical Thinking (selected by Lenny Rachitsky as a top course 🔥). Get 25% off the course with promo code suprainsider (auto-applied here).


This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit suprainsider.substack.com

What happens to the PM, the designer, and the engineer when a single person with a great idea can take it all the way to market on their own?
In this episode of Supra Insider, Ben Erez sits down with Tomer Cohen, former Chief Product Officer at LinkedIn, where he spent 14 years and led the company’s product transformation through the AI era. Tomer unpacks the full stack builder mindset - not as a job title, but as a fundamental rethinking of how product development works when the bloated, process-heavy model of the last two decades gets collapsed back down to its original building blocks: idea, build, ship.
They explore how Tomer actually rolled this out at LinkedIn across thousands of people, including replacing the traditional APM program with one where candidates submit a working product instead of a resume, and where the final interview is building something end to end in real time. He breaks down the three archetypes he sees emerging - system builders, full stack builders, and specialists - and what he observed clearly separating top performers from the rest: not their tooling, but their judgment. The conversation closes with a Q&A covering how to navigate interpersonal friction as roles blur, when you actually need a specialist to step in, and what skills will remain stubbornly human no matter what.
If you’re a PM, designer, or engineer trying to understand where your role is actually heading and what to do about it right now, trying to think clearly about how to future-proof your career without falling for hype, or curious how a CPO at one of the biggest tech companies in the world actually tried to operationalize this shift - this episode is for you.
All episodes of the podcast are also available on Spotify, Apple and YouTube.
New to the pod? Subscribe below to get the next episode in your inbox 👇
Want company-specific interview intel?
If you’re preparing for PM interviews at Anthropic, OpenAI, Google, Stripe, DoorDash, Uber, or Figma… we think you’re going to love Insider Loops. You can read about it in Ben’s LI post, Marc’s LI post or check out the site directly below.
In this episode, we covered the following topics:
* (03:18) What the full stack builder mindset actually is and why now
* (04:00) How product development became a “massive behemoth” of process and org complexity
* (06:15) Collapsing the stack back to its original building blocks: idea to market
* (09:05) The GPT-4 moment: how Tomer’s team saw it coming and rewrote the LinkedIn roadmap overnight
* (13:00) How Tomer rolled this out at LinkedIn — from executive leadership first to the APB program
* (16:35) The new hiring process: submit a working product instead of a resume
* (20:07) What actually separated great candidates from good ones: judgment, not tooling
* (25:42) The three archetypes: system builders, full stack builders, and specialists
* (30:00) Your role is changing whether you want it to or not — and what to do about it
* (33:11) Why top talent at LinkedIn leaned in and everyone else didn’t — the growth mindset framing
* (035:55) The leadership hackathon: mandatory vibe coding for LinkedIn’s top R&D leaders
* (38:10) The role of human mentorship when AI can teach you almost anything
* (43:48) Q&A: when do you actually need a specialist to step in?
* (46:33) Q&A: navigating interpersonal friction as roles and lanes blur
* And more!
Links:
* Tomer Cohen’s Podcast: https://www.linkedin.com/newsletters/building-one-7160766300821069824/
* Tomer Cohen: https://www.linkedin.com/in/tomercohen/
* Marc: https://www.linkedin.com/in/marcbaselga/
* Ben: https://www.linkedin.com/in/benerez/
If you’re a new subscriber, we encourage you to check out some of the recent episodes you might have missed:
To support the podcast, please check out the links below:
* Check out Ben’s AI Practice Copilot and top-rated Maven course: Self-Paced PM Interview Bootcamp: Product Sense & Analytical Thinking (selected by Lenny Rachitsky as a top course 🔥). Get 25% off the course with promo code suprainsider (auto-applied here).
* Supra has teamed up with Maven to bring you something special – courses that our own community members have personally curated. And because you’re part of the Supra family, you can get up to 35% in some of these handpicked selections with code SUPRAxMAVEN.


This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit suprainsider.substack.com