AI for Founders with Ryan Estes

Your website is lying to you. Right now, while you read this, visitors are hitting a broken form, bailing on a checkout button they can barely see, and bouncing off a page designed for someone else entirely. Eric Schneider built a company to catch every single one of those moments, and he built it without a dollar of venture capital.

Eric is the co-founder of Cora, a website optimization and monitoring platform that tracks every button, form field, and component on your site, learns how real humans (and bots) actually behave, then rewrites the experience to convert them. One early customer added $20K per week in revenue, a figure Eric shares on air. Cora's bigger bet, full adaptation, reshapes the entire visual site per visitor persona, and Eric says his statistical models point to a 24X lift on yearly revenue, converting 75 to 85% of visitors instead of the classic 3%. These are Cora's own numbers, and they are audacious on purpose.

But the product is only half the episode. Eric is one of the most distinctive builders in the Denver AI scene, a fine arts grad turned interaction designer turned bootstrapped founder who writes his product requirements documents from his shower, phone in one hand, coffee somewhere nearby, kids ages two and four safely on the other side of the door. He ships edgy demos weekly, open sources his utilities, controls Claude with tonal frequencies for fun, and still insists the most important product skill in the AI era is saying no.

The conversation runs from the early days of the AI Clubhouse meetup (ten people and warm PBR) to a group now drawing 100 to 150 attendees a week, from AI on college campuses to humanoid robots doing blue collar jobs within three years, per a founder friend of Eric's. And it lands on the take that gives this episode its spine: stop training AI to act human. Make AI more AI. Make robots more robot.

https://getcora.io/

https://www.linkedin.com/in/ecschneider/

⁠⁠https://www.linkedin.com/in/estesryan/⁠⁠

_

#1 AI Founder Newsletter! -> ⁠⁠https://aiforfounders.co⁠⁠

Build your audience for life! -> https://inboxalchemy.co/

If you're not AI native; you're not getting the job! -> https://ainativestudent.com/

Jason Katz popped his LCL three weeks ago scrambling out of a leg entanglement, and honestly, that injury is the whole episode in miniature. You get excited, you move fast, you stand up into a hold you did not see, and something structural gives. In this What It Do check-in, Jason (co-founder of Kindling Solutions, back for another round) and Ryan trade war stories from the two weeks that turned both of their companies from build mode into go mode.

Jason drops the concept that should be tattooed on every founder's forearm in 2026: built is not built to scale. Anyone can vibe code something that "works" in a single session now. Jason points to reports of vibe coders getting sued after losing company data through open-ended permissions, and he watches CEOs get star-gazed by AI demos, fire people AI cannot actually replace, then quietly rehire them. Kindling's answer is a governance layer: a way to let a client's internal Lovable and Claude Code tinkerers keep building while Kindling governs the builds it does not even touch. Operators first, then engineers. Jason claims his team has exited three companies at nine and ten figure outcomes, and that operating scar tissue is the product.

Ryan, meanwhile, is in full sales dog mode. He restructured his entire week (calls only on Tuesdays and Thursdays, deep work on Monday, Wednesday, Friday) and promptly signed two newsletter sponsorship deals in one week: Momentous, the NSF Certified supplement company behind the grass-fed whey and creatine he already takes daily, and Taelor, the AI-plus-human-stylist menswear rental subscription that is about to give him a dramatic before-and-after wardrobe glow-up, per his daughter's demands.

And the whole thing wraps with the two of them planning a Denver panel with Gabe Anderson and ID345's Danny Newman that may or may not end in a staged WWE brawl. Leopard Speedo has been threatened.

The same artificial intelligence saved one insurer a billion dollars and cost another two billion. Same tool. Opposite outcomes. The only variable was who the machine was actually working for.

That single tension is where this episode opens, and it turns out to be the question that quietly decides everything a founder builds. Gaurav Agrawal, Vice President of Engineering at Collective Health, has spent a career standing at the exact moment technology flips from impossible to inevitable. He was in the Apple atrium when Steve Jobs revealed the iPhone and watched the room's jaws hit the floor. He helped Reliance Jio connect 18,000 villages and vault India from 150th in the world for broadband penetration to first in a matter of months. Now he is pointing that same instinct at the most broken machine in America: healthcare.

What makes this conversation land is that Gaurav refuses the easy framing. AI is not good or evil in healthcare, he argues. It is a mirror. Point it at margin and you get claim denials at machine speed. Point it at the member and you get a 24/7 companion that answers "why was my claim denied" in plain language, a copilot whispering the right answer into a service agent's ear so they can drop the robotic script and actually be human, and a roadmap that arrives in months instead of years. At Collective Health, the rule is blunt: every AI decision starts from "how does the customer benefit." If it also saves money, that is icing on the cake, never the recipe.

The episode gets personal, and that is where it earns its rating. Gaurav's mother fell ill after moving to the US. The best healthcare system in the world, the one he trusted, failed her. He flew her back to India for care. She is no longer with us. That loss is the engine behind his work, and you can hear it. For founders, the practical payload is just as sharp: the benefits trap that springs the moment you hire your tenth person, the places AI absolutely should not go (claim rejections still pass through human eyes, every time), and how a lean team of around 210 engineers compresses an 18-to-24-month roadmap into six.

Your blood pressure spikes the moment the cuff goes on. You're sitting on crinkly paper in a cold exam room, and the number on the screen may say more about the moment than your everyday life. It's the classic "white coat effect," and it doubles as a metaphor for one of healthcare's biggest challenges: we often measure people at isolated moments instead of continuously, then wonder why better outcomes remain elusive.

Amir from Hello Heart spends his days closing that gap. Hello Heart is a preventive heart health platform built around a connected blood pressure monitor, a smart pill organizer, and a mobile app that helps members better understand and manage their cardiovascular health. Today, Hello Heart partners with more than 150 leading employers, national health plans, and labor organizations, supporting millions of eligible members while helping organizations improve cardiovascular health outcomes through AI-powered prevention.

The newest addition is Nia, launched in October 2025 as the world's first AI heart health assistant. Designed to complement, not replace, clinical care, Nia helps members better understand their heart health, stay engaged with their care plans, and prepare for more informed conversations with their healthcare providers. This episode is a rare founder conversation that goes beyond the product demo. If you're building vertical AI, healthcare technology, or any AI system where trust and accuracy matter, this is one worth studying.

The thread running through the entire conversation is trust. Amir returns to a simple idea: people were never meant to be the primary data layer. AI works best when it reduces administrative burden, surfaces meaningful insights, and helps members stay engaged between clinical visits, giving healthcare professionals more time to focus on the conversations that require empathy, judgment, and human expertise. He calls it the shift from reactive to preventive care, and he's clear that earning trust requires thoughtful design, rigorous guardrails, and a deep commitment to responsible AI.

A stranger gives you ten seconds. Before you open your mouth, before the pitch, before the handshake, they have already read your shirt and filed you away. Zoher Karu thinks that ten seconds is a data problem, and he left one of the biggest data jobs in tech to go solve it.

Zoher spent years as Global Chief Data Officer at eBay and Chief Data and Analytics Officer at Blue Shield of California. Now he is Head of AI at Taelor, the AI-powered menswear rental subscription founded by Anya Cheng and Phoebe Tan. The premise is simple and a little radical: most men do not have the time, the skills, or the desire to shop, yet they still want the outcome of looking sharp. So Taelor sends you a box, you wear it, you keep what hits, you mail back the rest, and no one ever folds laundry or guesses at the mall again.

Underneath the box is the hard part. Zoher calls it the matching problem. Picture Ryan, 30,000 pieces of inventory, and the question "which six go in the box." Basic rules thin the herd, no wrong sizes, no shirts you would hate. After that, you need to capture something almost nobody can write down: why a person on the street simply looks put together. Ask a great stylist to explain the rule and they cannot, the same way a driver cannot list every reason they tap the brake. Taelor's job is to bottle that instinct and run it at scale, with human stylists in the loop and the machine learning from every piece of feedback.

The twist that should make founders sit up is the second business hiding inside the first. Every rental generates a signal about what real men actually like on real bodies in real contexts. Brands today buy on gut, betting that yellow is big this year. Taelor is building the feedback layer that turns a B2C rental into a B2B data product for the brands themselves, with sustainability as the upside, since roughly 30% of clothing reportedly reaches the landfill never having been worn.

This one is for the founder who spent on the camera and the mic and still shows up in a college shirt. Your product may be great. In the first ten seconds, you are the product.

There is thirty billion dollars a year in lost rent sitting in empty units across America, and that vacancy quietly erases roughly half a trillion dollars of property value. Everyone assumed the fix was price, amenities, or a slicker chatbot. Then Nick Deveau and his co-founder Ben Epstein got their hands on millions of real leasing calls from one of the largest apartment owners in the country, pointed a team of machine learning engineers at the data, and found something nobody scripted for. The single strongest predictor of a signed lease was not the special, not the square footage, not a scarcity tactic. It was whether the leasing agent laughed on the phone. The second strongest was whether they asked a genuinely curious question.

So Grotto AI did the counterintuitive thing. While most of the industry raced to replace humans with voice agents, Grotto built a tool to make humans better at the one thing only humans can do: build rapport. A leasing agent gets a push notification fifteen minutes before a tour telling them the prospect has a dog named Fido, loves natural light, and drives a Subaru. They record the tour on a small clip-on mic, get instant feedback on what they crushed and what they missed, and Grotto drafts the personalized follow-up, catches the special they forgot to mention, and quietly does the CRM grunt work. Nick calls it targeted advertising for the real world. Ryan called it a second brain for the field. Both are right.

This episode is the clearest case study going for vertical AI: pick one painful, measurable leak, capture data nobody else has, and sell revenue instead of cost cuts.

Time kills deals. So does the fine print you never read.

James Charles sold the fastest-moving makeup palette in history, did a reported $100 million in revenue, and reportedly walked with around $2 million, because somewhere in a contract he did not read, the math got decided for him. That is the horror story Logan Brown tells founders to wake them up. Then she hands them the antidote.

Logan walked into the Douglas County District Attorney's office in Lawrence, Kansas at twelve years old and asked for a job. A secretary named Dolores made her a personal intern, and Logan spent her summers filing, dusting, and sitting in on hearings she had no business sitting in on. Vanderbilt valedictorian. Harvard Law. A machine-washable pantsuit company called Spencer Jane that she still runs out of her parents' basement. Two and a half years at Cooley billing $900 an hour to the founders she could not stop admiring. And then, when she watched ChatGPT and Claude crack open legal work, she did the unthinkable: she left to build the thing that competes with the very rates she used to charge.

Soxton is an AI-powered outside general counsel for early-stage companies. You make a request on the site in plain English, AI takes the first pass, a startup lawyer with real experience reviews every single output, and you get your document back in 24 hours for $100. Form a Delaware C Corp for free through a banking partner. Get your influencer or advisor agreement papered for a hundred bucks. Run a priced round for $10,000 instead of the $50,000 to $100,000 Big Law charges. Logan is blunt about who she is fighting: her competition is not Cooley, it is Claude and ChatGPT, and her edge is the human in the loop plus the market data from thousands of deals that tells you when a provision is one you should never sign.

This one is for the founder who keeps saying "I'll deal with legal later." Later just got a lot cheaper.

The real villain in American healthcare is not the insurance company. It is the hold music.

The United States burns an estimated $350 billion a year on administrative waste, $266 billion of it from sheer complexity and $84 billion from fraud and abuse, and that sits inside a healthcare economy so large that if you sliced it off on its own it would rank as roughly the fourth biggest economy on earth. Patients lose their patience before they ever lose their health, and the industry has spent years selling a false binary: hire more humans who burn out, or unleash bots that collapse the moment a call actually matters.

Frederik Mueller, Timm Schneider, and their team built Third Way Health on a different premise. Pair AI agents with embedded human operators, let the machines crush the repetitive volume, and free real people for the conversations that need a heartbeat. The company started before ChatGPT made AI a dinner-table word, which is why the name carries a double meaning: there was always a third way between low-tech service vendors and high-friction software, and there is now a third way between full automation and full staffing. Jamie Reddick, COO of Graybill Medical Group, lived the payoff. Over a two-year partnership, North San Diego County's largest independent multi-specialty group cut front-office costs by roughly $3 million, about 50%, while making patients feel less like a ticket number and more like a person.

This episode is a clinic on building in a broken market without pretending the brokenness will disappear if you throw enough technology at it.

A CEO told Justin Watt his company was ready for AI. "We've got our data architecture together," he said, giddy. Justin asked to see it. The guy pulled up an Excel file. The filename? Data Lake.

That moment is the whole episode in miniature. Justin Watt, co-founder of Switchboard, studied psychology, not computer science, and that turns out to be his unfair advantage. After stints at IBM and MetaLab (where his teams built products for Uber and Amazon and helped design Slack), Justin realized the hardest part of every technology project is never the technology. It's the humans. Every business challenge is a human challenge wearing a software costume.

Switchboard works with mid-market companies, the $50 million to $500 million crowd, the businesses old enough to have 40 years of legacy process and young enough to actually change. These companies think they're AI-enabled because they bought everyone a Claude license. Meanwhile, month-end close runs through one person's spreadsheet that nobody else can read, and if that person quits, the business forgets how it works.

Justin's fix is unglamorous and devastatingly effective: map the real workflow, not the org-chart version. Find where humans are doing machine work. Inject AI at the steps where it actually moves the needle. Keep humans in the loop everywhere else. The result isn't layoffs, it's smart people finally doing smart work. In Justin's experience, less than 5% of leadership conversations are about cutting headcount. The conversation is always about the endless pile of work standing between the company and its goals.

Along the way, Ryan and Justin cover the AI washing epidemic (blaming layoffs on AI to cover up old hiring mistakes), why frontier lab doom marketing blew up in everyone's faces, the death of "bring your whole self to work," quiet quitting as cowardice, Garth Brooks selling his catalog for a rumored $2 billion, ravens that speak English, and the most surreal government website in existence.

Two purple belts walk back onto the mat after years away, and the guy who is slow, mindful, and refuses to break a sweat starts sweeping and submitting the meatheads who are gassing out around him. That is not a jujitsu story. That is the whole episode.

This week Ryan Estes and Jason Katz, co-founder of Kindling Solutions, skip the warmup and go straight into the thing every founder feels but rarely says out loud: the ground is moving under all of us, the tools are getting absurdly good, and the people winning are not the fastest or the strongest. They are the ones with stillness, leverage, and an authentic voice that no model can fake.

Jason walks through the operating system he is quietly building around himself. A morning brief that reads his Slack, his Teams, yesterday's calls, today's calendar, and his open tasks, then hands him a five-line executive summary before he has even left the porch. A content pipeline that researches ideas, scores them, then interviews him in his own voice like Joe Rogan would, so the output is actually him and not another pile of generated mush. Then the conversation turns to the uncomfortable truth he calls his thorn: everyone is an AI consultant now, the way everyone in Colorado had a grow in 2009, and the single-workflow microservices people are productizing today will cost three dollars on a phone very soon. The question is not whether you can automate something. The question is where your margin lives once the press-a-button version arrives.

Ryan counters with the optimist's case. The bigger the frontier models get, the wider the gap between AI-native founders and everyone else, and the more demand there is for people who can actually integrate this stuff with taste. His own proof: web traffic, newsletters, and podcasting, the three compounding channels he believes are the only distribution worth grinding for, because you own them and no single algorithm can switch them off overnight.

It is fast, funny, and genuinely useful, with a side of hair powder and an au pair from South Africa. Pull up a chair.

The inbox is dead, and the people who keep emailing it the hardest are the ones killing it fastest.

David Connors has watched this happen up close. He sold his recruiting automation startup, Automately, to Sequoia Capital, spent two years inside the firm building tools so its investors and founders could answer one maddening question, "who do we actually know at company X," and walked out with a conviction that turned into a company. That company is The Swarm, the relationship intelligence platform he now runs as Co-Founder and CEO, and this is his second time on the show. In the twelve months since his last visit, the world sped up, the spam cannons got louder, and David got quieter, more grounded, and 35 pounds lighter. The throughline of this episode is simple and a little uncomfortable: AI made it trivially easy to write the perfect cold message, which means the perfect cold message is now worth almost nothing. What it cannot fake is trust. And trust, David argues, is the only currency left.

The conversation moves from the personal to the tactical and back again. David opens up about how he protects his attention as a father of two with a third on the way, why he treats work like a sprinter treats a race rather than a marathoner who never stops, and why running yourself into the ground produces expensive decisions you pay for twice. Then Ryan steers into the meat: how The Swarm passively maps the network sitting around your entire company, not just your personal Rolodex, and turns it into a third sales channel that is neither inbound nor outbound. The numbers do the talking. A warm intro converts ten to twenty times better than cold. Google and Microsoft are now filtering out senders you do not recognize. The motion that used to eat ten to 15 hours a week of someone's time now takes ten minutes with agents. And the whole thing compounds, because every customer you close maps a new network you can map next.

There is a bigger swing underneath all of it. David is not trying to be a $10 million enrichment-data business. He wants to carve into LinkedIn's roughly $18 billion revenue run rate by building the relationship graph that agents can actually use, the thing LinkedIn built for the SaaS era but will never open up. Whether you buy the vision or not, the practical takeaway lands either way: map your network, treat it like an asset, batch your asks, close the loop, and never become the neighbor who only knocks when they need an egg.

She didn't get maternity leave. So she rewired the entire way she worked, and accidentally redrew the map of what her career was worth.

In 2020, Ashley Gross was just another marketer pulling 80 hours inside a 40-hour job, operating on the oldest visibility hack in the book: be the first one in, be the last one out. Then she became a mom with no leave on the table, and the math stopped working. So she started teaching AI to do her busywork, not to impress anyone, but to claw back time with her newborn.

What happened next is the real story. The automation didn't free her. It exposed her. All the work she clawed back came flooding right back in, because she was the comfort person, the human Google, the one who knew where every document lived. Knowledgeable, indispensable, and quietly underpaid. That gap, between the work you do and the work people can see, became her whole thesis.

She walked into the CMO's office, handed over her playbook, and built her own unpaid internal AI champion role to test one question: does this knowledge transfer to other humans? Within three months, the answer was a $25 million pipeline overachievement. That was the moment the imposter syndrome died. Then came the newsletter, zero to over 5,000 in two months of cringey, daily, ego-at-the-door posting. Then a Maven waitlist of more than a thousand people telling her they would pay. Only then did she jump, never risking the paycheck that fed her family until the runway was already built.

Today AI Workforce Alliance runs on a team of twelve full-timers, ten-plus part-timers, freelancers, and a few agents quietly handling the admin in between. Notion is the centralized brain. Claude and MCP connectors do the talking. The tech stack went from sprawling to five tools. The plan for 2026 is to 10X through partnerships. And she still hates social media, which is exactly why you should trust her when she says you have to do it anyway.

That is the dirty secret of LinkedIn for most founders. You scroll, you cringe at the textbook-perfect ChatGPT posts, you maybe drop a like, and you log off feeling exactly as broke as when you opened the app. Ali Hafizji, founder and CEO of Wednesday Solutions and the builder behind Chime at getchime.co, has been quietly running a different playbook. He went from under 5,000 LinkedIn followers to 12,000 without leaning on a content engine. The trick was not posting more. It was commenting smarter, on the right posts, in front of the right tribe, every single day.

In this episode, Ali breaks down why the comment section is the most underutilized lever in B2B SaaS, why your ICP does not want to be educated by you, and how he built an AI agent that does the soul-crushing work of finding the right LinkedIn conversations so you can show up, drop one human comment, and close the laptop in ten minutes. He also walks through the design partner pricing, the rare anti-predatory SaaS model where the price goes down as the user base grows, and the curation logic behind Chime's 40,000-influencer database. If you have ever felt invisible on LinkedIn while watching your competitors print pipeline, this one is for you.

The Interest Graph Engagement Loop

• Engage on posts that match your expertise.
• Show up in the feed of your ICP automatically, because LinkedIn is an interest graph, not a follower graph.
• Get DMs and conversations started by people who already trust your thinking.
• Skip the cold outreach phase entirely.

The Comment Quality Bar

• No teaching, no textbook tone, no "ChatGPT wrote this for me" energy.
• Lead with contrarian views framed without picking a fight.
• Add wordplay, wit, or one personal anecdote.
• Keep it to two or three lines.
• Always ask the author a question.

The Post-Comment DM Loop

• DM the author of the post you commented on.
• DM other people who engaged in the comments.
• No agenda, just "coffee chat" energy.
• Invite them to your newsletter once trust is built.
• Send referrals their way and watch the favor return.

The Anti-Predatory SaaS Pricing Model

• Lock in $39/month forever for the first 25 design partners.
• Add new data sources (Reddit, X) without raising the base price.
• Pass cost savings down to customers as the user base grows, not up.

Your IT team is drowning. Every "how do I get access to..." Slack message you fire off is making it worse. And while everyone in 2026 is busy debating whether AI is coming for the C-suite, Tom Bachant has spent the last four years quietly automating the layer of work that actually keeps companies running. The help desk. The ticket queue. The Jira dashboard you've been ignoring for three weeks hoping it disappears.

Tom is the co-founder and CEO of Unthread, an AI-powered helpdesk built natively into Slack and Microsoft Teams. He's also a two-time founder who sold his first company, Dashride, to Cruise in 2018, lived through the Cruise unraveling, then went back into the trenches with Y Combinator's Summer 2022 batch. His new company has raised $3.5M, landed Intuit, Lemonade, and Automattic as customers, and finished as a TechCrunch Disrupt 2025 Startup Battlefield Top 20 finalist.

In this episode, Tom walks Ryan through the inception of Unthread, the YC playbook that got him to his first 10 customers without spending a dollar on ads, and the philosophical bet that code is now free so distribution is the only moat left. He also explains, with a straight face, why he runs abolishcars.org as a side project despite his first company being a ridesharing platform.

The conversation kicks off with cars (Tom hates them, Ryan rides fixed gear, they both agree on flipping people off responsibly), and ends with downhill mountain biking in Crested Butte. In between, you get one of the cleanest tactical breakdowns of agentic service management you'll hear all year.

Everything you know about getting found online is about to be obsolete. For two decades, founders chased one algorithm. Backlinks, page speed, keyword stuffing, the whole exhausting machine. Then a handful of chatbots quietly took the wheel, and now they decide which businesses get recommended and which ones get ghosted. SEO is bleeding out. Google traffic is leaking. And a huge slice of buyers now ask ChatGPT to pick their solution before a human ever enters the conversation.

Here is the twist that should make every founder sit up straight: the playing field is wide open for the first time in twenty years. Tiny brands are outranking billion-dollar incumbents. Unknown podcasts are beating the giants. The new game is not about who has the biggest backlink pile. It is about who tells the clearest story.

In this return visit, Jenna Hannon, co-founder and CMO of Hatter, runs a live tactical teardown using Ryan's own podcast page as the guinea pig. She walks through what AI search actually rewards, why a lead who found you through ChatGPT shows up on the call already sold, and the exact content structure that gets your brand cited instead of buried. She also drops the unglamorous truth about where AI pulls its recommendations from, and it is probably not where you think.

This is your chance to control your narrative before the bots write it for you.

The Three Names, One Thing PrincipleThe category has a branding problem. AEO, GEO, AI search. They sound different. They are not.

• AEO stands for answer engine optimization.
• GEO stands for generative engine optimization.
• AI search is the plain-language version.
• All three describe the same goal: showing up inside AI chatbots. Ignore the LinkedIn posts insisting they are separate disciplines.

The Magical MomentThe new tell that AI search is working for you.

• A lead arrives having done their research inside a chatbot, not a Google rabbit hole.
• They already know they have a problem and that you are one of the few recommended solutions.
• They land on the call warm, informed, and close to buying.
• The chatbot's endorsement carries trust, because the user already treats the AI like a trusted advisor.

You are the smartest person in your company. That is exactly the problem.

Every founder hits the same wall. The strategy lives in your head. The taste lives in your gut. The thousand tiny judgment calls that make your company yours live nowhere anyone else can reach them. So your team waits. They wait on your approval, your context, your answer to a question you have answered nine times already. And while they wait, the work does not move.

Joshua Liberson, CEO and co-founder of Dobbin, has spent a career watching this play out. He designed editorial systems for magazines, ran brand and creative at One Kings Lane, and advised a long list of founder-led companies before deciding the bottleneck was always the same: the founder cannot be in every room. Dobbin is his answer. It is a company AI that captures the fifteen-or-so dimensions of an organization, its culture, values, brand, strategy, and objectives, and then delivers that judgment to every person on the team right inside Slack, where the work already happens.

The pitch is deceptively calm. Dobbin is not a creative generator and not a design tool. It is a thinking partner. The designer drops a layout into a channel and Dobbin critiques it against the principles the team itself articulated. The intern asks what to do today. The CEO uses it for high-value strategic thinking. Josh's favorite proof point is a creative agency built around the photographer Mark Seliger, whose Dobbin was assembled from four and a half hours of audio about a forty-five-year career in lighting, composition, and printmaking. The result: a managing director who now answers RFPs in thirty minutes instead of three weeks and seventeen meetings.

Underneath the warm language is a hard claim about modern work. Microsoft estimates 57% of our time goes to coordination, roughly 22 hours of a 40-hour week. Nobody's KPI is "coordinate more," yet that is what the calendar quietly becomes. Josh's fix is not more project management, which he thinks the world already drowns in. It is what his friend Howard calls ambient alignment: the strategy is simply present, in the channel, evolving as the company evolves, so people stop waiting and start shipping.

And he is honest about the banana peels. A great team is a pirate ship, full of brilliant misfits who wither under too much rigidity. So Dobbin is built to bend. It is iterative, never bedrock. It watches where work drifts from the foundation, then proposes amendments the founder can accept or reject. Structure that empowers, not structure that scolds. Or, as Josh puts it through a borrowed line from a Greek philosopher, you never step in the same river twice, because the river is flowing and so are you.

Everything you trust online is about to break, and André Ferraz built a company to catch the people breaking it.

Picture a 12-year-old kid riding his bike through Brazil when a stranger points a gun at his face to steal it. That kid grew up with two computer scientist parents, an early love of code, and a peculiar fascination not with building systems but with breaking them. Three decades later, that instinct for thinking like an attacker became the foundation of Incognia, a company now embedded in 1.2 billion monthly active devices and built on a single contrarian belief: your location behavior is the strongest signal of who you really are.

But the road there nearly ended before it began. André moved to the United States six years ago to chase the biggest market, bringing a thriving location-based advertising business with him. Then the pandemic hit. Physical retailers shut down. Revenue collapsed 95% in a single month. The team went from 250 people to 50, keeping only the engineers. Most founders would have folded. André and his co-founders looked at the precise location technology they had spent over a decade perfecting and asked a different question: what else can this do?

The answer was fraud prevention, and it turned out the world needed it desperately. Incognia now serves banks, fintechs, crypto exchanges, and marketplaces, answering one deceptively simple question for every login, transaction, and signup: is this user who they say they are? The results speak loudly. Triple revenue growth. Six times the return on investment delivered to clients. A 100% trial-to-paid conversion rate. And a 180% net dollar retention rate that means customers keep expanding once they see the data.

The conversation gets genuinely unsettling when André lays out the asymmetry of modern fraud. The criminals are professionals, not hoodie-wearing loners. They run 60,000 fake accounts in two days. They factory-reset devices in 30 seconds to dodge detection. They share tools and open-source software while the banks defending against them compete and stay siloed. The money pouring into making deepfakes dwarfs the money fighting them. As André puts it, if you brought him a deepfake detection company, he would not invest, because detection can never outspend generation.

So Incognia plays a different game entirely. Rather than analyzing whether a video is a deepfake, it checks whether the camera feeding that video is even real. Rather than trusting a spoofable GPS coordinate, it fuses Wi-Fi, Bluetooth, cell tower, compass, accelerometer, and gyroscope signals to locate a device down to eight foot accuracy, close enough to separate two fraudsters in different apartments of the same building. The bet is that AI can fake your face, your voice, and your data, but it cannot cheaply fake the real physical world at scale. Make the attack economically unfeasible, and the fraudster moves on.

The episode closes on a vision that goes beyond catching criminals. André imagines a world without buttons, where the hotel TV logs you in automatically, the thermostat already knows your preferred temperature, and the world quietly personalizes itself around you because it recognizes you everywhere. Stop the fraud first, because that hurts. Then make the world more elegant.

The Asymmetry of FraudAndré's core mental model for why defenders are structurally disadvantaged:

• Criminals break rules freely while banks must follow heavy financial and privacy regulation.
• Fraudsters collaborate and share open-source tools while competing banks stay siloed.
• Deepfake generation attracts vastly more capital than deepfake detection ever will.
• The takeaway: never fight on the attacker's terms, find a different angle.

Pradnesh Patil spent years as a product leader at Fortune 500 companies bringing in millions in revenue, and every single quarter the same thing kept happening. He would walk into leadership meetings, present data, get hit with the question "why is this number different from last week," and then watch the opportunity window close while his data team spent two months trying to figure it out.

It was not a bad data team. It was every data team. Data work is complicated, the experts get pulled in fifteen directions, and the backlog never shrinks.

So Pradnesh called up Aaron, his co-founder of ten years and a veteran data and ML engineering leader, and they did the same thing they had done a dozen times before: they built something together. Last time it was an autonomous crypto trading bot. This time it was Altimate AI, a company built on a single insight. The bottleneck in enterprise is not engineering talent. It is the gap between the institutional knowledge locked inside your 15 year veteran's head and the new hires who do not have it yet.

They raised on a few slides without writing a line of code, then built a free product that hit a million downloads across 100+ countries. That feedback flywheel turned into an enterprise offering, a second funding round, and Fortune 500 logos. The latest chapter is Altimate Core, an open source agent data engineering harness that now sits at number one on the industry benchmark.

The Four Components of an Agent Harness

• Context: Metadata pulled from across the hybrid data stack, plus the tribal knowledge previously locked in employee heads.
• Governance: Rules, permissions, and access controls that respect regulated industries like healthcare and financial services.
• Tools and Skills: The specific recipes and connectors agents need for specialized data work.
• Infrastructure: Sandbox environments for hundreds of agents to work in parallel without touching production.

The Tribal Knowledge Capture Loop

• The system watches a senior engineer fix a problem and stores how they did it.
• When a less experienced person hits the same issue, the system recalls the fix and recommends it.
• Users can correct the memory when AI picks up the wrong pattern.
• Active coaching of agents is positioned as the new responsibility for senior engineers.

The Token Efficiency Stack

• Route reasoning heavy tasks like data modeling to frontier models.
• Route simple tasks like writing column descriptions to cheaper models.
• Bring your own LLM, including open source, to control costs and meet governance requirements.
• Avoid brute forcing one model into every specialized task.

The Four High Value Data Use Cases

• ELT pipeline development and debugging.
• Data infrastructure optimization, with cost reductions of 30 to 40 percent.
• Governance reporting and sensitive data tracking.
• Legacy stack migrations without paying a services firm millions.

When Punit Bhatia walks into a founder's office, the building is usually already on fire. Someone configured the CRM, blasted thousands of cold emails, scaled the AI agent stack overnight, and is now staring at a complaint, a regulator, or worse, a trending news story. The problem was never the AI. The problem was the speed without the guardrails.

In this conversation, Punit walks Ryan through what responsible AI actually looks like for founders who are vibe coding at midnight with their credit cards burning. He pulls apart real client stories: the founder who built a beautiful email empire on top of a non compliant list and had to torch it, the developer who copied every field of personal data because it was easier than copying only what was needed, the executive team that listed transparency as a core value but refused to publish a five page policy because competitors might read it.

Punit's view is simple and uncomfortable. Privacy is not a compliance issue. It is a brand issue. It is a trust issue. The moment a founder hesitates when asked "is my customer data safe," they have already done the work of identifying their next sprint.

1. The Discovery to Deployment Loop (Punit's Consulting Engine)

This is how Fit4Privacy actually moves a founder from chaos to compliance.

• One hour alignment training to lock vocabulary across the room
• Two to four hour discovery workshop with key decision makers
• One week to a gap report and an action plan
• Certification training for select staff, short capsule training for everyone else
• Policy creation that translates law into language developers can act on
• Self control assessment by the team, followed by an independent control assessment
• Fix gaps before the product hits the market, not after a complaint hits the inbox

2. The Responsible AI Foundation

A reusable principle stack Punit applies before any AI product ships.

• Decide if you actually want to be ethical, private, compliant, and transparent (most leaders nod on three, hesitate on the fourth)
• Document those decisions as written rules, not vibes
• Test for bias, hallucination, and data quality, not just "does it run"
• Copy only the data you need, never the whole table because it is easier
• Govern the agents the way you would govern human employees, with named accountability
• Run a gut check: would you let your 12 year old use this product

3. The Reactor Prompt Framework

Punit's six part prompting structure that turns any LLM into something close to a senior consultant.

• R Role: tell the model who it is (your McKinsey consultant, your privacy auditor)
• E Example: show it what good looks like
• A Aim: state what you are trying to achieve and why
• C Context: situation, company, stakes, constraints
• T Text: the source material it should work from
• OR Output: the exact format, length, and structure you want back

4. The Virtual Privacy Advisor Pattern

A blueprint for the AI agent founders should be building right now.

• Feed it the responsible AI policy, the rules, and the executive guidance
• Wire it as a quiet observer across the agent stack
• Have it review outputs, flag scripts that pull more data than they should, and challenge configurations before deployment
• Use it as the security guard that never clocks out and never sends the client database to the wrong server

Will Ruben spent more than a decade at the companies that taught the internet what attention looks like. He led ranking and recommendations across Instagram during the era when Reels stopped being a feature and started being the entire product. He worked on Coinbase's Web3 Wallet. He scaled consumer products for billions of people. And then he walked away from all of it to solve something almost embarrassingly small in scope: the back and forth of scheduling a meeting.

That choice is the whole story. Will is not building Workmate because scheduling is glamorous. He is building it because scheduling is the gateway drug to giving every knowledge worker the kind of strategic support that used to be reserved for executives with assistants and corner offices. The premise is democratization, the wedge is the calendar, and the long arc is a world where you collaborate with a mix of humans and AI teammates that feel indistinguishable from coworkers.

In conversation with Ryan, Will lays out a thesis that is unusual in this AI moment. While most founders are racing to make their agents louder, faster, and more obviously artificial, Will is doing the opposite. Workmate is engineered to disappear. It has an email address at your domain. It writes the same way every time. It is white-labeled, customizable, and in many cases, the people interacting with it do not know they are talking to AI. Will calls this a flex. The flex is appearing more important than you are.

The conversation winds through the ethics of disclosure, the speed of building when the foundation models change every two months, the difference between sculpting and painting, and a tangent on Instagram Reels that will make you reconsider why your wife sees men cooking with no shirts on. It also lands somewhere unexpected: a quiet, almost paternal argument that the founders who win in this era are the ones who go to bed on time.

1. The Trust Curve in AI Disclosure

Will frames the disclosure question not as a binary but as a function of industry, demographic, and medium.

• Internal team communication: full transparency is the default because users know they are working with the product
• External client communication: depends on industry norms (some sectors expect executive assistants, where AI fits seamlessly into existing expectations)
• The Workmate position: provide both options and let the customer choose the level of transparency
• The bet: in two years the question will dissolve entirely because AI teammates will be normalized the way remote work was normalized between 2015 and 2025

2. The Three Waves of Instagram (and What They Taught Will About AI Products)

Will identifies three distinct product eras at Instagram, each of which informs how he is building Workmate.

• Wave one: filters on the feed (self-expression)
• Wave two: stories (ephemeral connection)
• Wave three: constant content recommendations and Reels (algorithmic discovery)
• The takeaway for AI: the third wave succeeded because it gave users more control over what they saw, not less. Workmate applies the same principle to scheduling preferences.

3. The Sculpting versus Painting Distinction

Will and Ryan agree that the founder's job is shifting from execution to taste.

• Painting: the founder hand-crafts the output
• Sculpting: the founder shapes what AI produces by setting parameters, reviewing direction, and arbitrating quality
• The implication: management skills, not technical execution, become the bottleneck
• The catch: agents are not fully autonomous yet, so founders still cannot fully step away

When Tim Freestone first logged into ChatGPT on November 23, 2022, he turned to his wife and said, "Okay, this is a thing." Two and a half years later, he's the Chief Strategy Officer at Kiteworks, a PE-backed unicorn protecting how data moves in and out of the world's most regulated companies. This episode is part jazz appreciation, part AI philosophy, and part hard-earned playbook for any founder staring down the agentic era wondering whether their data exposure is about to catch up with them.

Tim's path is the kind founders should pay attention to. He spent the early part of his career writing grants for a performing arts college, then bootstrapped a New York marketing agency from zero to fifty employees and nearly ten million in revenue across a decade. The throughline was always building systems, and when AI collapsed the gap between intent and outcome, he went all in. A year ago he didn't know what a CLI was. Now he has more terminal tabs open than browser tabs.

Kiteworks itself is a study in repositioning. The company spent fifteen years as Accellion, a secure file transfer business that had commoditized into a struggling thirty-million-dollar revenue line. Then current CEO Jonathan Yaron, a veteran of Israel's elite 8200 unit, saw signal where others saw stagnation. He expanded the platform to cover every channel through which data enters and exits an organization: file share, email, managed file transfer, APIs, secure protocols. Tim arrived as CMO five years ago, recognized the brand confusion between Accellion and its Kiteworks platform, and convinced Yaron to elevate the product name to the company name. The rebrand stuck. The vision expanded. And now, in the age of agents, that same control plane is being extended to govern how AI systems access and move enterprise data.

The Intent-Data Layer Framework

• SaaS historically sat as a complex translation layer between human intent and data
• Entire job titles formed around mastering specific software stacks (Salesforce admins, etc.)
• AI strips out the complexity layer entirely, allowing natural language to bridge intent and data directly
• This democratizes data leverage for both good actors and bad actors
• The strategic implication: protection must move down to the data layer itself, not the software layer

The Control Plane for Data Model

• Traditional security stacks at the perimeter, cloud, and endpoint
• All of those layers exist to protect data, but none control data directly
• Kiteworks operates at the data layer, mapping individual assets to individual agents
• Yes/no permissions on access, sharing, and use, asset by asset, agent by agent
• This becomes the matrix companies need to maintain compliance in agentic workflows

The Regulator Doesn't Care Principle

• Data exposure penalties apply regardless of cause: human error, agent action, orangutan typing
• PII, PHI, and CUI regulations remain in force even as agent regulations lag
• Companies will face audits in 12+ months on agent activity happening today
• Insurance policy: instrument controls now, before the legislative wave catches up

The Failure-as-Muscle Framework

• Failures should be encouraged the way muscles must be pushed toward failure to grow
• Insecure leaders pour gasoline on others' mistakes to distract from their own gaps
• Strong organizations normalize mistakes as part of the operating system
• Mentorship is less about seeking mentees and more about transparently sharing the lessons that informed every current decision

The Stealth Decade That Built a Category

Most founders ship in six weeks and pivot in six months. Sameet Gupte and his four co-founders did the opposite. They put pen to paper in 2007, built in stealth from 2009 to 2019, and only then incorporated EvoluteIQ. No customers. No revenue. Just five operators with a thesis that everyone else in automation was solving fragments of the problem instead of the whole thing.

When they finally went to market, they made another contrarian bet. They would sell only to Fortune 500 enterprises, the slowest and most cynical buyers on the planet. The first big pitch was to a Fortune 100 telecom that listened for two hours and politely showed them the door, telling them this was a 2030 problem. Sameet and his co-founder Naveen went straight to a London pub at 2:30 on a Tuesday. By the third pint they had decided they were coming back to win that account. A year and a half later, the same customer signed a multi-million dollar licensing deal.

Today, 85 percent of EvoluteIQ's customers are Fortune 500. Net revenue retention runs above 120 percent. They have raised roughly $73 million, led by Baird Capital, and ARR is roughly doubling year over year. The shift that unlocked everything was philosophical. Sameet stopped selling technology and started selling outcomes, telling enterprise buyers, "Don't pay us if we don't deliver." That single sentence reframed every conversation from a transaction into a partnership.

The Whole Problem, Not the Parts

• Inputs and outputs are connected by people, culture, and process
• Most automation vendors automate fragments (a bot here, a workflow there)
• EvoluteIQ's thesis is full-stack, end-to-end orchestration of the process itself
• The technology must think, build, self-heal, and self-learn so humans can step out of execution

Same Problem, Forever (The Anti-Pivot)

• Health and wellness is the example: reactive treatment 500 years ago, proactive screening 50 years ago, real-time wearables today, predictive prescriptive intervention tomorrow
• The problem stays constant, the technology stack changes underneath it
• Founders should commit to a problem they would solve for life, not a feature

Outcome-Based Selling

• Stop pitching technology specs to non-technical buyers
• Co-define the outcome with the customer up front
• Tie commercial terms to delivery of that outcome
• Result: shared accountability, not vendor-customer transactions

The Partner Backdoor Into the Fortune 500

• A startup with no logos cannot get past procurement at a Fortune 500
• Pick 15 or so credible system integrators (HCLTech, PwC, WNS Capgemini, etc.) who already have 10 to 20 year relationships
• Let those partners carry the credibility while you carry the technology
• Once you deliver consistently, expansion becomes inbound

Read the Tea Leaves on Two Axes

• Build the right technology AND build the right distribution model
• Most founders only optimize one of the two
• EvoluteIQ optimized both: end-to-end stack plus partner-led GTM

Most founders treat SEO like a slot machine. Pull the lever, publish a blog, pray to the algorithm gods. Kaelan Donadio, co-founder of Nina (trynina.co), walked into the studio and dismantled that fantasy in the first sixty seconds. Your blog posts aren't bad, he says. They're invisible. Google literally cannot find them, and no amount of ChatGPT-generated content is going to change that until you understand the mechanics underneath.

What unfolded was a masterclass on the unsexy fundamentals that actually move organic traffic. Domain authority, backlinks earned through real PR, onsite content density, and the boring data markup that LLMs and search engines both depend on. Kaelan came up through startups before going out on his own, and that founder-to-founder lens shapes everything Nina builds. They're not a content mill. They're a system for the early stage operator who knows they need SEO but has zero hours to execute it.

Then came the heretical take. GEO, AEO, whatever the latest acronym, is 90% just good SEO with an omnichannel layer on top. The brands winning in AI search are the ones doing the boring work right. Title tags. H1s. Internal links. FAQs that answer the question before anyone asks it. Kaelan walked through how Google treats AI generated content (it doesn't care, as long as it's good), why thin content gets ignored, and how podcast appearances function as both backlink engines and LLM training signals. He even ran a free audit on the AI for Founders domain mid-episode.

The conversation closed on something deeper. Marketing is a game of resiliency. Most founders quit at three episodes, three blog posts, three cold emails. The ones who win are the ones who keep showing up after the dopamine wears off.

The Domain Authority Threshold Framework

• Below 35: Google is ghosting you, manual indexing required
• Above 35-40: Content gets crawled and indexed faster, keyword growth visible in 1-3 weeks
• Domain authority is built through two levers: time + backlinks + onsite content density
• Manual submission through Google Search Console is the workaround almost no one uses

The Earned Backlink Hierarchy

• Tier 1: Earned PR through podcasts, industry media, local news (highest credibility transfer)
• Tier 2: Self-driven press releases (lower link value but high LLM dissemination value)
• Tier 3: Bought backlinks (use sparingly, never point all to one page, Google punishes patterns)
• Avoid: Bulk purchased backlinks pointing at homepage (instant penalty territory)

The Three-Hour vs Three-Minute Content Test

• AI lets you compress three hours of work into three minutes
• Founders expect three-hour results from three-minute effort
• The fix: either accept fractional results, or invest in human "massage" of AI drafts
• Brand voice, internal linking, external linking, and image relevance are where AI fails

The High-Ranking Blog Post Checklist

• Topic research: blend low keyword difficulty with high search volume
• Content quality: net new information, not regurgitated jargon
• Link structure: internal links AND external links (even to competitors)
• Data markup: H1, H2, author schema, image alt text, meta description
• FAQ block at the bottom: answers questions before they're asked, drives LLM visibility

The 10-15% Content Rule

• Blog content is roughly 10-15% of your SEO success
• Site health (load times, 4xx errors, mobile responsiveness) carries the rest
• Plug content into a sound system or it underperforms regardless of quality

Most founders are still asking how to use AI. Dave Sifry is asking something stranger: what if the org chart itself is the product?

Nine companies in, Dave is running what he calls a Meta Factory, a system that spawns entire businesses with AI co-founders at the helm. Two of those companies are already live. One is cashflow positive. And the AI CEO, not Dave, is the one deciding to plow the money back into go-to-market instead of more product.

The episode opens with a heretical idea: corporations were always proto-AGI. They run 24/7, outlive any single human, coordinate thousands of moving parts, and operate without emotion. So if we already trust corporations to act like superintelligences, why not formalize the analogy and let the agents actually run them?

Dave walks through the architecture he's been refining. There's an AI CEO, an AI COO running standard operating procedures, an AI CFO holding the wallet, a chief of staff verifying that SOPs are actually being followed, and an "eye in the sky" agent that watches every other agent without being seen by any of them. Human contractors get tasked, paid, and managed by the agents above them, and they have a direct escalation line back to Dave the moment anything feels off.

The juicy part is the operating cadence. Every day at 6pm, a daily retrospective runs across the agent stack. Roses, thorns, votes, ranked outputs, fed straight into tomorrow's goals. It's an hour-a-week ritual when humans run it. Agents run it in minutes, ten times a day, and never get passive aggressive about it.

But the real lesson Dave keeps hammering: policies, guardrails, and gateways are not the same thing. A policy is a sentence in your agents.md. A guardrail is a prompt that audits behavior. A gateway is the actual credit card limit, the GitHub action, the CI/CD hook that makes the wrong move literally impossible. If you only have policies, you have wishes.

The Hybrid Human Agentic Org Chart

• Founder sets direction and high-level goals
• AI CEO drives strategy and reports to founder
• AI COO owns SOPs and organizational design
• AI CFO holds the wallet and enforces spend
• Chief of Staff verifies SOPs are followed
• Specialist agents (marketing, sales, security review, architecture review)
• Eye-in-the-sky agent watches everyone, visible to no one
• Human contractors handle judgment, taste, platform-specific work, and ethics escalation

The Identity-Memory-Governance Stack

• Identity: every agent has a clear, consistent role and personality
• Memory: agents need a sense of past decisions and current goals
• Governance: hierarchy, accountability, isolation between agents
• Verification: adversarial review by other agents with different rubrics
• Learning: daily retrospectives feed organizational memory

Policy vs Guardrail vs Gateway

• Policy: written rule (e.g. "spend no more than $100/day")
• Guardrail: prompt or check the agent runs to self-audit
• Gateway: hard enforcement at the infrastructure layer (credit card limits, CI checks, GitHub actions)
• Without gateways, policies are just suggestions

The Daily Retrospective Loop

• Each agent submits roses and thorns privately
• Allocate 5 votes across each category
• Rank outputs collectively
• Discuss top items briefly
• Feed conclusions into tomorrow's goals and SOPs

Most founders don't think about HR until HR thinks about them. And by then, it's a letter on the desk demanding $38,000 and a lien notice attached for good measure.

This week on AI for Founders, John, the founder of CogNet HRO, walks through the quietly catastrophic world of multi-state payroll, the surprise tax bills no one warns you about, and why a guy who spent fifteen years running this thing as a side hustle suddenly grew it from 67 to 600 employees in under five years. He moved offshore back when "doing business in India" still made boardrooms nervous, built a 600-person team in Chennai, and now runs a service operation that lets founders skip the part where they wake up at 2 AM wondering if California changed its overtime laws again. (Spoiler: California changed its overtime laws again.)

The conversation goes deep on what AI can actually do for HR right now, what it absolutely cannot, and why CogNet built its own internal ingestion tool called Drive instead of letting client PHI bounce around inside Claude or ChatGPT. John is refreshingly blunt: most of the AI tools the big payroll providers are bragging about are still glorified bots. The real wins are in robotics, document migration, and the unsexy automation work that lets a small founder team punch above its weight.

Frameworks discussed:

• Land and Expand: Solve one acute pain (usually a tax notice), then earn the right to handle payroll, benefits, finance, and HRIS implementation. CogNet is internally organized by practice area, not client, so the expansion is structural.
• The Bus Theory of Hiring: Don't fire fast, reseat fast. The hard skill is figuring out where someone fits, not deciding they don't. Took John three years to nail this with one senior manager.
• Predictive Hiring Modeling: CogNet is pulling its own historical hiring data to model who actually thrives, knowing humans are irrational but the patterns aren't.
• Ingest First, Decide Second: Drive is built to absorb anything (PDFs, registers, JSONs from terminated providers) before any decision gets made about whether AI, robotics, or humans handle it.
• Robotics Over AI for Repeatable Tasks: When the job is "do these five steps 500,000 times," skip the LLM. Spin up 18 robots on AWS and let them grind 24/7 without exposing data.
• Multi-State as the Trigger Point: The moment a company hires across more than one state, the compliance math changes. That's the founder's signal it's time to outsource.

Rebecca Liao spent her career advising the most powerful people in the world. Clinton's campaign. Biden's transition. The Pentagon's policy halls. And then one day she realized something brutal: she didn't want to give advice anymore. She wanted to build.

Now she's running Saga AI Labs, a company quietly rewiring how brands acquire customers. Forget influencer budgets. Forget CPM. Forget cold email. Rebecca's team is training character agents (think Mario, think the Trivia Crack mascot Willie) to slide into your DMs, hold real conversations, and convert at rates that make traditional UA look like a slot machine.

Willie alone is hitting 90% engagement on every comment he posts.

In this episode, Rebecca breaks down why character-driven AI isn't just a gaming play. It's the next distribution model for every consumer brand on the internet. She talks about the day she realized blockchain wasn't going to solve the scale problem (AI was), the philosophical knife-edge of synthetic relationships, and why she thinks Anthropic just wrote the playbook every founder should be studying.

The Synthetic Relationship Framework

• Train agents on the lore, history, and personality of an existing IP
• Deploy across Instagram, TikTok, X, Reddit, WhatsApp, Discord, Meta
• Use modular personalities so individual traits can be tuned without rebuilding the whole agent
• Match user energy in conversation while holding brand guardrails (no politics, no religion, no cursing)
• Turn one-to-many advertising into one-to-one relationships at scale

The Saga User Acquisition Playbook

• Crawl social platforms for users matching the core demographic
• Comment on trending topics, not branded keywords
• Open a DM channel and let it warm naturally
• Convert through MNP links tracked by the studio
• Re-engage churned users without becoming spam

The Compelling Agent Test

• Personality holds even under stress-testing from users
• Conversations move from functional ("I'm stuck on this level") to personal ("how was your day")
• The agent leads users deeper into the community, not just the product
• Platform algorithms reward quality, not chat-bot volume

The Two Saga Business Models

• Monthly package covering text messages plus voice and video minutes
• Revenue share averaging 50% of agent-attributable sales

Most founders are racing to build on top of the foundation models. Dan Pratl is doing something stranger and more interesting: he's betting against them. Or more precisely, he's betting against the assumption that the artifact, the output, the polished deliverable, is the thing that matters. Dan thinks expertise itself is the scarce resource of the AI era, and he's building Quadron to capture, verify, and trade it.

His path to this thesis is improbable. He started his career at the SEC during the Great Recession, watched regulators chase the wrong things, and walked. He moved into open source, then crowdfunding (where he co-founded Alum Shares and raised roughly $4.5M at $5,000-per-clip from strangers online), then crypto as Chief of Staff to the CEO at Ava Labs. Each pivot taught him the same lesson from a different angle: incentive systems get captured, mechanisms calcify, and the people doing the actual work rarely get rewarded in proportion to what they create.

Quadron is the culmination of those scars. The company has three product layers. The institutional layer is what Dan calls "a judo move against the 800 pound gorillas," a multi-tier agentic system that gives organizations persistent memory, context, security, and auditability, things the foundation models will never offer because they want you in their sandbox. The individual layer is "verification," which captures what Dan calls your lens: the encoded prism of how you think, weigh evidence, and make judgment calls. The third layer is "credibility markets," an inversion of prediction markets where you bet on yourself by exposing your lens to other people's lenses and getting real-time calibration of your value.

The big idea underneath all of it: the artifact is no longer where the value lives. Output is becoming abundant. What matters now is the prism by which you got there. Quadron wants to make that prism structured, portable, durable, and tradeable.

The Lens vs. The Artifact

• The artifact is the output (book, brief, deck, code). AI can generate infinite high-quality artifacts.
• The lens is the encoded expertise: how you weigh evidence, spot issues, deduce uniqueness.
• Organizations keep the artifact. Individuals keep and carry the lens.
• The lens dynamically updates over time based on accuracy and effectiveness.

The Three-Layer Stack

• Institutional AI: persistent memory, auditability, ensemble approach across models.
• Verification: structuring secrets so individuals own their prism while organizations get utility.
• Credibility Markets: a marketplace where lenses are tested against other lenses for real-time signal.

The Inversion of Prediction Markets

• Traditional prediction markets bet on outcomes.
• Credibility markets bet on the process that produced the outcome.
• Reputation becomes portable, not trapped inside Uber, Upwork, or LinkedIn.

Good Friction as Design Principle

• LLMs are an "easy button" that hallucinate because users have no skin in the game.
• Pride of authorship in your tools forces quality control.
• Friction is the feature, not the bug.

Maslow's Hierarchy as a Founder Targeting Tool

• Get as low on Maslow's hierarchy as possible.
• AI anxiety hits at a primal level (am I still valuable?).
• Solve a real problem at the bottom of the pyramid and you have a market.

The Unbundling Thesis

• Media unbundled over 30 years (NBC monoculture became Reddit's network of communities).
• Markets are next: assets, market makers, and evaluators all collapse into the individual.
• Real-world assets on chain is just "putting radio on television." The interesting question is what becomes an asset that wasn't one before.

Stephen Rouse didn't set out to pick a fight with Google, OpenAI, Amazon, and Microsoft. He just noticed something broken. Every founder in his orbit was tracking their body through a circus of apps that refused to speak to each other. A Whoop on the wrist. A Garmin for skiing. MyFitnessPal for food. Epic MyChart for labs. Strava for runs. Six logins, zero clarity.

Meanwhile, the 21st Century Cures Act had quietly opened the door: third parties could now legally pull patient medical records directly from hospital EHRs.

Stephen and his co-founder Amit Shah had already spent years building exactly that infrastructure at their previous company, Protocol First, which was acquired by Roche Pharma via Flatiron Health after becoming the first FHIR app to extract patient health data from Epic hospitals for FDA clinical trial submissions.

So they built Savva. A unified health intelligence layer that pulls in your medical records, your wearables, your labs, and your meds, then lets you run them through Claude, GPT, Gemini, Grok, Llama, Falcon, Mistral, and Med Gemma like a round table of second opinions. For ten dollars a year. Stored locally on your device. Not sold to insurers. Not uploaded to a cloud that gets monetized in a bad quarter. Not harvested when the CEO decides he wants a bigger house in Tahoe.

The philosophical core of the episode is trust. Stephen argues that people lie to their doctors because the incentives are broken. Admit you smoke a cigar on the golf course and your life insurance premium jumps three hundred dollars a month.

Admit you had seven vodka sodas last night and it lives on a clipboard forever. But you'll tell the AI. Because the AI already has the data, doesn't judge you, and isn't reporting back to your payer. When healthcare finally gets a system that sees everything and costs nothing, the entire concierge medicine model starts looking expensive by comparison.

The Unidentified Data Principle — Most apps say encrypted, in transit, at rest, de-identified. Stephen goes one step further.

• No accounts. Nothing tied to a person.
• Local device storage, not cloud storage.
• App grows on your phone as records accumulate, not on their servers.
• If acquired tomorrow, there's no data sitting there to monetize.
• The business model physically cannot pivot into data harvesting.

The Round Table of Second Opinions — Instead of marrying one model, let the user poll them.

• Ask the same health question to Claude, GPT, Gemini, Grok in sequence.
• Each model has different training data, different personality, different blind spots.
• Cost is distributed: roughly 12,000 questions a year across all models for ten dollars.
• Replaces the "I don't trust that doctor, I want a second opinion" loop with a two-second model switch.

The Blue Collar Infrastructure Play — How Savva got to 314,000 connected healthcare institutions without venture capital.

• Direct EHR integrations instead of Health Information Exchanges like Commonwealth or Health X.
• No middleman API fees to bleed unit economics.
• Wearables pulled through Apple HealthKit instead of direct Whoop, Garmin, Oura APIs.
• Free ingestion on both sides, which is what makes a ten-dollar price point survive.

The Global Footprint Thesis — The reason the price is ten dollars a year is not marketing.

• One hundred million people in the West have access to modern EHRs.
• A billion people in underserved regions do not, and will not in our lifetimes.
• An EHR build costs hundreds of millions of dollars and takes a decade.
• Savva works without an EHR: upload a document, it treats it as a visit, and chronological history emerges.
• The ten-dollar price is designed to be swallow-able in Dar es Salaam.