START

The biggest hedge funds have entire teams building AI. Most asset managers are on their own.But closing that gap isn't about buying more AI. It's about a problem most people get wrong.You can test code by running it.You can't test an investment by buying it.Software engineering and investing aren't the same problemIn software, we've moved from autocomplete to agents to long-running agent swarms - each step climbing another level of abstractionIn investing, human judgment still has to exist at the object levelSo the challenge was never deploying agentsThe challenge is giving them the context to be usefulEvery source of knowledgeInternal and externalStructured and unstructuredConnected to the people making decisions.That's what Ian M.J. McInnis and his co-founder Ben Finch built at WithAI (YC P26)Not another AI toolA customized environment where agents can work the way investors actually do.Ian knows the problem firsthand. Before co-founding WithAI, he worked at the center of AI and investing at Bridgewater, studying how AI would reshape markets and knowledge work.His conclusion after trying the alternative:"The DIY approach ends in pain"The future isn't simply managing swarms of agentsIt's knowing what to delegateWhat to automateAnd where human judgment remains essentialThat's why WithAI focuses on amplifying investors rather than replacing themResearch fasterMonitor more companiesSurface information that would otherwise be missed"AI for asset managers, built with you": Security, infrastructure, and white-glove service for harnessing the latest agents🎙️ Ian McInnis, CEO & Co-Founder, WithAI on Fondo START pod‍00:20 Why AI adoption in investing requires more than simply deploying agents00:35 Building a customized AI operating environment for asset managers01:12 Making hedge fund-grade AI capabilities accessible to smaller firms01:55 From Princeton mathematics to a career in investing02:35 Becoming one of Bridgewater's internal AI specialists03:25 The origin story of WithAI and partnering with a longtime collaborator04:35 Identifying the missing tools investors would need in an AI-driven future05:05 Lessons from an initial YC rejection and what changed06:15 Why AI agents are creating immediate value for knowledge workers07:20 The fundamental difference between software engineering and investing08:35 How to balance AI automation with human judgment09:50 Where investors can safely delegate work to AI agents11:00 Measuring AI-driven investment outcomes and client value creation12:15 The ideal asset manager profile for WithAI today13:10 Opportunities and excesses emerging in AI infrastructure investing14:00 Thoughts on SpaceX, xAI, and technology valuations15:00 The future of AI adoption across institutional investingCheck out www.withai.co

Shipping isn't the bottleneck anymore. Attention is.Maanav Agrawal kept seeing the same pattern.Some of the coolest products weren't getting the attention they deservedMeanwhile, companies that consistently shared what they were building were generating inboundHe'd lived it himselfHe saw it in other industries like real estateHe saw it across his YC batch and among founders generallyThe problem wasn't always the product.It was making sure people knew the product existed.So he and Co-Founder Jason Zhan built Memoir (YC P26)A marketer with the soul of an engineer. Memoir connects to your product source of truth and understands what changed, why it matters, who should hear about it, and how to tell the story.A new feature shipsA PR worth talking about gets mergedA meaningful update goes liveMemoir turns it into:→ Social posts → Blogs → Changelogs → Demo videos → Customer updates → Launch contentAll in your company's voice.They built it for themselves: Engineers who love buildingBut, like most founders, don't always have time to build in public.Now they don't have to.The work speaks for itself - automaticallyMemoir turns product velocity into market velocity🎙️ Maanav Agrawal, CEO & Co-Founder, Memoir on Fondo START 00:53 "Shipping is not the bottleneck, attention is."01:19 How Memoir connects directly to GitHub and monitors product changes01:38 Deciding which product updates are important enough to share publicly02:03 Keeping founders in control with approval before publishing02:40 Why some of the best products never get the attention they deserve03:20  The challenge facing startups without dedicated marketing teams03:49 Why engineering teams move faster than marketing teams can keep up05:17 The original YC idea: developer infrastructure and latency optimization05:58 The realization that marketing was the biggest bottleneck in real estate06:18 The insight behind Memoir: unify engineering velocity and marketing velocity‍Check out trymemoir.ai‍

In 2022, San Francisco leased five million square feet of office spaceThe entire yearIn Q1 2026 alone: more than four millionAt the time of this recording (Apr 2026) tenant demand reached 7.5 million square feetThe highest ever recorded.OpenAI has accumulated nearly a million square feet in Mission BayCoinbase, Nvidia, and the Warriors all landed at Mission Rock (It's effectively full - premium space there is almost impossible to get)Reuben Torenberg has brokered SF startup office deals through every major cycle since 2016The 3% vacancy boom and record rentsThe 38% vacancy COVID collapseAnd the AI-driven resurgence happening nowWe're probably still early in this cycleThe AI wave showed you can do more with lessAnd that's changing how founders think about space🎙️ Reuben Torenberg, Senior Vice President, CBRE on Fondo START00:36 Why brokering compounds the longer you do it01:08 SF's pre-COVID boom, then the jump to 38% vacancy02:18 How AI restarted SF office demand02:42 Where we are in the cycle, and why it isn't the top03:25 The leasing numbers that signal a real shift04:34 Why startups are leasing differently now05:34 How distressed buildings changed hands at a discount06:11 How landlords regained leverage07:38 What landlords actually check before they'll sign you09:48 Why plug-and-play subleases win with startups12:18 Advice for founders hunting their first office13:01 Treating the office as a recruiting tool, not a status symbollearn more at cbre.com

A single glass of water holds about 20 hours of breathingLeo Kankkunen is building the gear that pulls it outTankless diving technology that generates breathable oxygen directly from water. It sounds impossible at first. Then you hear the physics...Then you realize the implications stretch far past divingWater is hydrogen and oxygen. Apply a DC current, swing away the electrons binding them, and you get the purest form of breathable gas.The idea started with a simple frustrationLeo is an electrical engineer and a diver. Every dive exposed the same limits: Bulky tanks, Logistics, Safety risks, Cost, and Operational complexity.Instead of accepting those constraints, he asked a question: Is there any way we could breathe like a fish?That sent him down a rabbit hole into electrolysisIf water already contains oxygen, why haul tanks at all? Why not generate breathable oxygen exactly where and when it's needed?Enter DAIVIN! (YC W26) - Tankless dive gear that generates oxygen directly from the water around you. No tanks to haul, fill, or run out of.But diving is just the wedgeAnywhere breathing depends on a tank, the tank sets the limit. Crisis zones. High altitude rescue. Anywhere time is human lives. Generate the oxygen on site, and there's no supply chain to depend on.🎙️ Leo Kankkunen, CEO & Founder, Daivin on Fondo START pod00:18 What Daivin is building and why oxygen tanks are the real problem00:43 One glass of water holds roughly 20 hours of breathing01:05 The science of splitting water into oxygen and hydrogen with a DC current01:38 The wearable system that draws in surrounding water and produces oxygen02:10 Leo's path from electrical engineer and diver to founder02:48 Where the tech stands today: commercial close, consumer further out03:15 Why removing oxygen logistics matters in crisis response and rescue03:38 The Netflix versus Blockbuster framing for oxygen infrastructure04:05 Why water beats gas tanks for breathing beyond Earth05:20 From a Slush side event in Finland to a YC interview within weeks06:25 Building in San Francisco versus Finland and compressing months into weeks07:15 Leo's advice for founders: build what genuinely interests youCheck out daivin.tech

Startups spend more time wiring their outbound tools together than actually talking to potential customersVerbiflow was built to change thatIt's Outbound Infrastructure for growth teamsOutbound used to be a stackOne tool for email.Another for LinkedIn.Something else for cold calls.Verbiflow is the system that runs your outbound, all in one placeRun sequences through email, LinkedIn, and cold callsHandles sendingManages repliesRuns follow-upsKeeps it all in syncBecause the real bottleneck isn't finding leadsIt's getting in front of the right people before momentum disappearsAnd in a world flooded with AI-generated outreach, they're betting on: better conversations, not more messagesFrom prospect to pipelineFrom outreach to conversationFrom disconnected tools to one outbound system🎙️ Kash Ambarani & Rishi Mahadevan, Co-Founders, Verbiflow on Fondo START pod‍00:57 Platforms for email, LinkedIn, and cold calling are completely separate02:08 We built an MCP that gets you live in 10 minutes03:25 Why AI "personalization" in email is making outbound worse04:18 Cold calling is the most underrated channel right now05:16 AI coaching and real-time objection handling on calls06:03 A targeted list is way better than trying to fill a giant list07:35 Try to get there in person and actually talk to them08:10 How they mapped every Series A-C startup in SF by scraping T&Cs‍Check out verbiflow.com

Most brands don't know if their marketing is working.They launch campaigns. Spend real money. Watch the dashboards. Then hope.Months later, a report arrives explaining what happened. The campaign is over. The budget is gone.The market has already moved.That's how brand tracking has worked for years.Henk and Harry are repeat founders with exits behind them - two decades building brand trackers for the world's largest brands on one side, a startup built and sold plus a market research career on the other. They'd watched the same problem from the inside for years.Tracking was expensive. Slow. Built around snapshots. Not how brands actually grow.So they built Timelaps: An AI-native brand tracker that continuously surveys thousands of real consumers and shows whether your marketing is moving the metrics that matter. Brand awareness. Consideration. Preference. Category ownership. Brand growth.Research-grade tracking. Updated continuously. Running in days instead of months. At a fifth of the cost of traditional tracking.From "What happened?" to "What's happening right now?"‍🎙️  Henk Pretorius & Harry Zhang, Co-Founders, Timelaps on Fondo START pod‍00:17 Building AI-native brand intelligence for modern brands00:45 Why brand tracking remains one of marketing's biggest pain points01:53 How two founders met at ODF and began validating ideas together02:42 Combining market research expertise with startup experience03:24 Landing customers from a prototype before building the full product03:46 Winning Product Hunt with an early demo04:35 Why the original customer problem never changed05:30 The limitations of traditional brand tracking and brand audits06:03 Moving from delayed reports to continuous insights06:40 The metrics that matter most in brand intelligence07:35 Making enterprise-grade brand tracking accessible to challenger brands11:43 Why San Francisco accelerates founder learning and execution12:47 The "Olympic Games of building companies" mindset‍Check out www.timelaps.io

The internet got flooded with AI content.Everything started sounding the same.Same hooks.Same cadence.Same “we’re not doing X — we’re doing Y.”Jameson Zaballos thinks that’s exactly why human taste matters more now.Napa is the team behind those viral startup memes you’ve probably already seen.But underneath the memes is a bigger thesis:The future of founder marketing belongs to people who still sound human.Not polished.Not optimized.Human.AI can generate infinite content.But it still averages the internet.The founders winning attention right now are the ones adding taste back into the loop.Write like you talk.Post when nobody else posts.Make your ICP feel seen.Usenapa.com🎙️ Jameson Zaballos, Co-Founder & CEO, Napa on Fondo START01:07 Why founders can't cut through the noise02:12 "If you outsource the opinion to Claude, you level the playing field"02:45 How AI writing quietly cheapens your brand04:15 Why human unpredictability is still the moat05:58 The overlooked advantage of posting on weekends08:42 Why memes work when they reflect real customer pain09:55 The return of long-form storytelling

Global markets were built for everyone. Access wasn't.Millions of traders outside the US watch Tesla move. Watch Nvidia earnings.Watch SpaceX rumors explode timelines.But when it's time to act? The door closes.Mochatrade is rebuilding that layer.A perpetual futures platform for US equities. Built for global traders first.Trade TSLA, NVDA, AAPL, and 50+ US stocksUp to 50x leverage24/7 marketsNo US brokerage account requiredDeposit in local currency. Trade in minutes. Stay self-custodial the entire time.Because the future of trading won't look like legacy brokerage apps.It'll be:BorderlessAlways-onCrypto-native infrastructure underneathSimple UX on topIndia already drives massive derivatives (60%) volume. But global traders still can't easily access leveraged US equities.That contradiction became the company."We have experienced the pain point ourselves" Parth and Chetan grew up in India watching this door stay closed.So they built the platform they wished existed.Not another broker. An access layer for the next generation of retail traders.🎙️ Parth Maheshwari & Chetan Manda, Co-Founders, Mochatrade on Fondo START pod‍00:57 Why global traders still can't access leveraged US stocks02:21 Turning INR deposits into seamless US stock trading03:27 Why India dominates 60% of global derivatives activity04:06 Trading pre-IPO companies like SpaceX through perps05:10 What perpetual futures actually are and why they're simpler06:27 Why perps fix the retail options problem07:45 From AI agents to financial infrastructure08:52 Leaving AI to rebuild global market access09:30 How three IIT friends and former roommates found the idea‍Check out www.mochatrade.com

The best AI companies aren't replacing bad software. They're giving it a login.Every software company for years has made the same pitch to manufacturers: clean APIs, migrate your stack, rip out your ERP.It never works. The ERP is the company brain. Years of data. You can't rip it out any more than you can rip out someone's memory.Nikolas Keller & Co-Founder Lukas Postulka figured out the obvious thing that everyone missedStop trying to replace the software. Give an AI employee a login instead.Walter (YC P26) signs into SAP, Microsoft Dynamics 365, and Oracle the way a new hire would. Ready to go from day 1.Works out of Teams and email like the rest of the team. Reads orders. Enters them. Places supplier POs. Catches pricing errors before they ship...Not another dashboard. Not another integration project. Not another rip and replace.A purchase order that used to take 15 minutes now takes a few seconds.The insight came from a $109M manufacturer who had a full-time employee doing nothing but typing purchase orders into an ERPNot because the work was valuable. Because the software demanded it.With Walter, your software stays. The manual work disappears.🎙️ Nikolas Keller, CEO & Co-Founder of Walter (YC P26) on Fondo START02:02 Growing up across Beijing, Zurich, Munich, and Singapore02:53 Walking into French restaurants asking for a job04:10 Why he walked away from the Michelin-star path04:48 Choosing startups because uncertainty was the point05:25 Flying to San Francisco with a duffel bag06:18 Getting rejected, then forcing his way into a startup07:06 Learning software engineering and customer discovery08:00 Meeting future co-founder Lukas Pistoor09:00 Lukas' bias toward action and unconventional journey10:16 Discovering the manufacturing workflow problem10:48 Selling Walter before the product existed11:26 Why AI employees work where ERP replacements failvisit www.walter.one to learn more

Relationships drive revenue.That's why more companies are investing in dinners, customer events, founder meetups, and community gatherings.But here's what happens after the event:You sourced the attendeesYou sent the invitesYou hosted the eventThen everyone goes homeNow you're digging through Slack messages, WhatsApp chats, notes apps, spreadsheets, and email threads trying to remember:Who showed up.Who you talked to.What was discussed.Who deserves a follow-up.Most of that context never makes it back into the business.The event created value.The workflow lost it.Pavoot is an AI event manager built for companies that run customer eventsIt helps teams find the right attendees and draft personalized invitationsIt shows who's in the room and why they're relevantIt lets teammates capture notes around each attendee and share context with each other in real timeAnd afterward, those conversations can flow back into the CRM instead of disappearing into someone's phone.The event was never the hard part.Keeping what it created is.Because relationships don't create value when they're madeThey create value when they're remembered🎙️ Gohar Tamrazyan, CEO & Co-Founder Pavoot (YC P26) on Fondo START00:22 Ana's journey from Brazil to AI research at ETH Zürich01:06 Gohar's chess background and Swiss national team experience01:40 Building a media management tool and discovering a larger problem02:45 Why companies are investing more heavily in in-person events03:14 The real goal behind customer events: relationships and outcomes04:05 Using AI to source attendees and build the right room04:21 Event recommendations based on attendee interests04:38 The challenge of remembering conversations after events05:00 Why teams still rely on Slack, WhatsApp, notes, and voice recordings05:30 Capturing attendee context and team notes in one place06:17 Launching Pavoot's Luma integration08:22 Which companies benefit most from AI-powered event managementlearn more at pavoot.com

Chris Bakke never pitched Elon. He just posted good ideas in public for nine months straight.Laskie was sourcing engineers on Twitter while everyone else lived on LinkedIn.Along the way Chris started tweeting his own takes on how Twitter should fix recruiting.Not as a pitch - just because he had strong opinions about what was broken.Elon liked one. Then followed him. For the next nine months Chris kept the takes sharp on purpose, knowing exactly who was watching.Then the DM came. A phone number and a note to call that Saturday. His wife was sure he was getting catfished. He called anyway, and Elon picked up on the first ring.Six minutes later it was an invite to dinner at the Tesla Fremont factory that week.Deal closed in 45 days - Elon's first acquisition at TwitterThen Twitter became X, X got acquired by xAI, xAI got acquired by SpaceX.Chris came out the other side holding SpaceX sharesNone of it happens if he builds quietly. The ideas did the outreach for him.🎙️ Chris Bakke, Founder with exits to @X @Indeed @Zillow to on Fondo START pod02:08 Building Laskie as a recruiting startup in the COVID remote-work wave03:24 The thesis: reverse-arbitrage engineering talent outside the US04:41 Why being in an exponentially growing space matters more than anything05:17 The Twitter takes that got Elon to follow, then DM06:33 What it's like getting a DM from the richest man in the world07:06 The Saturday phone call and the Tesla Fremont dinner08:29 Why you never take the acquirer's first number09:14 Using optionality as leverage10:48 Selling from a point of weakness as the hiring market collapsed11:22 Two years at X and xAI, and a deal around $50M12:05 The acquisition chain: shell company to Twitter to X to xAI13:19 SpaceX deal, and the shares he keptVisit www.chrisbakke.com to learn more

Teddy li's first version of Prepse did something reasonableIt connected to your call recordings, pulled out the data, and filled in your CRM. customers told him it was a nice add-on. he could have explained why they were wrong. he chose to find out why they were right.So he looked closer, and what the most interested customers actually wanted was underneath it.They didn't just want cleaner data. they wanted to use it to improve how their teams sold, to make their median reps perform like their top ones.The data was the input. The training was the point.He then worked with hundreds of enablement managers to learn what the best teams do differently, and built Prepse around the answer: reps run simulations of real buyer conversations before they have them for real.The real product was hiding one question deeper than the one he set out to answer.The useful question about ai in sales isn't what it can replace. It's what it can't. It can't decide how your company should handle a pricing objection, or what a good discovery call sounds like in your market.That judgment lives in a handful of your best people, and it's the scarce thing.He built Prepse on that premise. Instead of automating the seller, it takes the judgment your best people already have and turns it into something every rep can practice against. Your playbook, your objections, your definition of good, run as simulations a rep can repeat until the real call feels like the fourth take.The leaders set the bar. The software makes sure everyone can reach it.‍🎙️ Teddy Li, Co-Founder, Prepse on Fondo START pod‍00:00 Teddy introduces Prepse and what an AI enablement manager actually does01:00 The original product: pulling call-recording data into the CRM01:25 Why customers treated it as an add-on, and what they actually wanted instead01:45 Studying hundreds of enablement managers to model best-in-class teams02:05 Snagging the six-letter domain for ten bucks, and what Prepse stands for02:40 His previous company, Nofin, and going through YC03:05 Why a founder's skills carry over non-linearly between startups04:00 The two kinds of teams Prepse sells to04:30 The real goal: turning the median rep into a top rep05:35 His onboarding philosophy, getting users to the magic immediately06:30 What he tells founders about cutting onboarding friction07:00 Revealing product depth one layer at a time, like an onion‍Check out prepse.com

AI shouldn't replace therapists.It should help them spend more time being therapists.Not because AI won't get smarterBecause good therapy isn't just informationIt's trustIt's compassionIt's the human element.Most therapists spend only about half their time doing therapyThe other half?Running the businessNotes. Paperwork. Marketing. Insurance claims.Klarify automates the work around the session so therapists can focus on the work only they can doToday, more than 7,000 therapists use the platformThe result is clear: they spend less time running a business, more time doing the work only they can doMoody Abdul, Co-Founder & CEO of Klarify, on Fondo START Podcast01:53 Therapy changed my life02:22 A therapist asked to use my previous product02:31 Therapists spend much of their time on admin03:22 Surpassing 7,300 therapists on the platform03:46 Why AI won't replace therapists05:07 The real value hidden inside therapy notes05:28 Turning session insights into therapist marketing06:09 Building tools for independent therapists07:00 Fighting insurance claim denials with AI07:28 Why therapist SEO and client discovery matter08:17 The therapist Klarify serves today09:32 From solo therapists to 60-person clinics10:00 Solving therapist-client matchingLearn more at www.klarify.ca

Most companies don't realize they have a documentation problem until everyone already depends on it.Customers use it. New hires use it. Engineers use it.And when documentation falls out of date, the whole system starts working against itselfPeople stop trusting what they're readingTeams lose contextAnd nobody can fix it, because the problem is everywhere at onceThe longer it goes unfixed, the harder it is to untangleThat's the opportunity Manicule sawThey're building an AI-native DevRel company for developer tools - owning documentation, technical content, GEO, and distribution across social channels.The premise isn't that AI should replace expertise. It's that expertise should scale.Their AI agents audit and test at scale. Their humans own the architecture, the writing, and the creative direction.Every review improves the systemEvery project creates more context. Every iteration raises the barWhat started as a highly manual business - helping developer companies create better technical content - has become a scalable AI-native operationToday, Manicule works with some of the fastest-growing developer tool companies - including Supermemory, Greptile, and Reducto - has scaled fast over the last few months, and has more demand than it can take on.Not because they publish more content. Because they help developer companies create content developers actually trust.🎙️ Naman Bansal & Shreyans Jain Co-founders of Manicule (YC P26) on Fondo START00:57 What Manicule is building03:43 The Supermemory customer that changed everything04:47 Building startups together in high school05:55 Early pivots, recruiting agencies, and first revenue06:18 Applying to YC as teenagers07:24 Lessons from being the first engineer at Supermemory08:49 How DevRel became a growth engine09:54 Why long-form content beats viral marketing10:53 Growing Manicule 6× and crossing $60K MRR12:17 Why developer documentation is often broken13:19 How AI changes go-to-market without replacing quality14:42 The biggest misconception about GEO16:00 What founders misunderstand about getting into YC17:42 The hackathon that led to YC19:05 Turning a manual agency into an AI-native company21:33 Building toward $5M ARRlearn more at manicule.dev

A production issue shouldn't require a detectiveYet that's how most observability works todayMost observability tools just dump alerts. Duplicates, no context, and you still fix it yourselfNicolò Magnante  and his cofounder Arseniy Shishaev - who spent years building part of Datadog's metrics product - think that's backwardsSo they built Superlog. Observability that's meant never to be opened.A wizard scans your repo, installs proper OpenTelemetry, and runs daily to keep up as you shipWhen something breaks, it groups the errors into one incident, investigates with full context - logs, traces, recent deploys, past Slack threads - and drops a single mergeable PR in Slack.Merge itIgnore itOr open it in Claude Code and tweak itVendor-neutral, so you keep every log, trace, and metric - even if you leaveNot another notification. A fix.🎙️  Nicolò Magnante, CEO & Co-Founder, Superlog "Observability that fixes your bugs"02:00 Why existing observability tools create friction03:00 The YC pivot that led to Superlog05:22 Why observability must evolve beyond monitoring05:40 Installing observability with a single prompt06:14 Why startups are the ideal first customers10:01 Why the future is self-healing softwareLearn more at superlog.sh

AI agents can write code, analyze data, and pass the bar examThey can't order you a pizza because they don't have a phone numberThink about how much of the real world still runs on phone numbersSMS verificationDelivery coordinationAccount creationCalling a business...Without one, an agent is stuck behind glassManav Modi & cofounder Meet Modi built AgentPhone to fix thatGive your agent a phone number and it can suddenly actHe told his agent to order DoorDashIt created an account, talked to the delivery driver, made the paymentPizza showed up at his door. Y Combinator is a customer. After An engineer tried Twilio and got buried in compliance, AgentPhone had him live in a dayEvery person will soon have hundreds, maybe thousands of agents working for themThe hard part isn't making them smartIt's giving them the tools to touch the real world ‍🎙️ Manav Modi, CEO & Co-Founder, AgentPhone on Fondo START pod‍00:57 Why AI agents need phone numbers to interact with the real world01:24 The tools that move agents closer to human-level autonomy01:50 The emerging infrastructure stack powering AI agents02:18 Why payments may become a critical layer for autonomous agents02:42 An AI agent ordering DoorDash end-to-end without human involvement03:08 What happened when 100+ developers built on AgentPhone03:26 Processing 4,000+ calls and 5,000+ messages in a single day03:43 Unexpected real-world use cases discovered during the hackathon04:24 How phone numbers unlock authentication and action05:37 The vision of every person having hundreds or thousands of agents06:16 Building identity and authentication systems for AI agents07:31 The startup pivot that led to AgentPhone and ultimately YC‍Check out agentphone.ai

Flowscope’s agents can map an entire department in two weeks.Then they automate it.It’s an AI-native consulting firm where agents learn how a business actually runs - process by process - then automate the most manual work directly inside the company’s existing systemsNot in quarters. In days.Samuel kept seeing the same pattern every time a consulting firm walked into a large organization:Great strategyBeautiful slidesMinimal implementationThe recommendations were smart. The operational change rarely came.He also saw a second problem: Consulting fees were disconnected from outcomes. Clients paid upfront whether the recommendations worked or not.The incentives were broken by designSo he and his co-founder Javier built the thing consulting always promised but rarely delivered:Implementation.The pitch to clients is simple:Grow your company with the headcount you already have.🎙️ Samuel Mirpuri, Co-Founder, flowscope on Fondo START pod00:57 Why traditional consulting stopped at recommendations instead of implementation02:33 The disconnect between consulting fees and measurable outcomes03:18 How Flowscope maps entire enterprise departments in two weeks04:24 Why AI should augment human judgment instead of replacing people05:30 From aeronautical engineering classmates to AI-native founders06:02 Sam Altman’s $2M OpenAI token offer to YC startups07:00 Why AI tokens may become a new startup currency layer07:29 Customer, Customers, Customers‍Check out www.flowscope.com

99% of session replays are your app working exactly as expected (hopefully)The other 1% is where the friction hidesThe bugs.The failed onboarding flows.The conversion leaks.No team has time to watch them all.That's the problem Michael Egan and his co-founder kept hitting at their last company. They had analytics. They had session replays. They had ideas to improve conversion.The small fixes always lost to bigger priorities..So they pivoted and built CodeCanary: An AI product engineer that connects to GitHub, PostHog, and Slack. It watches your session replays, finds the UX friction, and opens pull requests with the fixAlways on. Always shipping.Not dashboardsNot alertsActual fixes‍🎙️ Michael Egan, CEO & Co-Founder, CodeCanary on Fondo START pod‍00:57 AI agents connected to your product analytics and codebase01:20 "You wake up to pull requests the next morning"03:41 Pivoting from enterprise mapping software04:42 Why nobody actually watches their session replays05:02 Surfacing the 1% of user behavior that matters05:47 Autonomous A/B testing tied directly to revenue06:37 Why this solves problems engineers never had time for07:30 "With every new user, your product gets a little better"08:30 How to know when it's time to pivot‍Check out www.codecanary.ai‍

Most cancer drugs target proteinsFinalDose thinks the real opportunity is one layer deeper: DNAJeff Liu and his team are building a programmable drug platform that uses genetic mutations to identify and selectively destroy cancer cellsCells without those mutations are not targetedInstead of creating a completely new drug for every cancer type, the platform is designed as a reusable system with different genetic instructionsOne chassis.Different disease targets.The broader bet: DNA is more deterministic than proteins, which could unlock therapeutic approaches traditional drugs struggle to reachSearch.Destroy."A programmable cell elimination platform."🎙️ Jeff Liu, Cofounder & CEO of FinalDose on START 00:11 Why proteins may be the wrong layer for cancer treatment01:14 Using DNA mutations as programmable kill signatures03:20 Building a reusable therapeutic platform instead of one-off drugs05:00 Why hard biotech problems require interdisciplinary teams06:00 Getting into YC after a 4 a.m. interview in Japan08:00 Sam Altman’s $2M AI token offer — and why FinalDose passed10:08 How AI-native biotech is accelerating programmable medicine12:00 Why the future of medicine may look more like engineering13:05 The regulatory bottleneck for programmable therapeuticsLearn more at finaldose.ai

The next big space race won’t be about going up.It’ll be about bringing things back down.After four years building satellites at Astranis, Payton Case realized something: as launch costs collapse, the bottleneck flips.There’s still no infrastructure for manufacturing products in space and returning them safely to Earth.So Dispatch is building it.The thesis is simple: gravity is an invisible constraint on manufacturing.Remove gravity and semiconductor defects drop. Pharmaceutical crystals become more stable. Biological structures that collapse on Earth can solidify in microgravity.To prove the concept, the team built a full-scale heat shield, drove into the Mojave Desert, and blasted it with a rocket engine at 14× expected re-entry force.It survived 6× what they needed.The long-term vision:Factories in orbit. Permanent industrial infrastructure in space.🎙️ Payton Case, Co-Founder & CEO of Dispatch, on Fondo START pod00:12 What Dispatch is building: reusable re-entry vehicles for orbital manufacturing01:16 The next big space race... will involve bringing things back down to Earth02:05 Expanding from re-entry vehicles to industrial space stations02:33 Mojave Desert heat shield testing03:48 Gravity is this invisible constraint on manufacturing04:06 Pharmaceutical crystal growth in microgravity04:53 Semiconductor manufacturing with lower defect rates06:39 Understanding the modern space infrastructure stack08:01 3D printing organs in microgravity09:37 The coolest thing to be done in space has not been thought of yetCheck out dispatch.space‍

Ansel Dias is using his robot factory to build more robot factoriesAutoFAB is a robotic desktop factory that prints, assembles, quality-checks, and packages hardware autonomouslyIt’s already being used to manufacture more of its own robotic armsThe insight behind it is simple: Prototyping is easyProduction is the completely different ballgame - and that’s where hardware startups dieFactories. QA. Packaging. Scaling production.That’s the bottleneck AutoFAB wants to eliminateThe long-term vision: AutoFABs making more AutoFABs🎙️ Ansel Dias, Founder & CEO, AutoFAB on Fondo START pod 00:18 AutoFAB is a robotic desktop factory00:33 It prints things, then assembles them00:41 It does quality checks and even packs shipping boxes01:03 Putting hardware into production is extremely difficult02:01 AutoFAB to make more AutoFABs02:14 Growth will become geometric02:37 Right now it's making more of its own robotic armsCheck out autofab.net to learn more

The most valuable thing in AI won't be generating answers.It'll be knowing which ones are right.Right now AI writes code, solves problems, produces proofs. But there's no way to guarantee any of it is correct. Pedro Nobre is building that guarantee.Cajal sits at the intersection of formal verification and AI. They use Lean, a language that lets you formalize a statement and derive a proof that's either correct or incorrect. Binary. No ambiguity.The hard part: the space of possible proofs is combinatorially large. Humans somehow navigate it with strange inductive biases. Machines couldn't keep up.Then reinforcement learning changed what's possible. AI can now iterate against an infinite source of reward: mathematical correctness itself.The thesis: create a superintelligent mathematician, and you solve most problems.They're already working with frontier AI labs. Starting in quantum computing and finance. Software verification and cryptography next.🎙️ Pedro Nobre, Co-Founder, Cajal on Fondo START Pod‍01:37 Formal verification explained - verifying whether software or mathematics is correct02:24 We need to make sure what AI outputs is correct03:07 Why mathematical proof search is combinatorially difficult03:42 How reinforcement learning is changing theorem proving04:11 Why AI is suddenly solving harder math problems04:28 We already have access to a superhuman mathematician04:46 The future of checking whether all mathematics is actually correct05:42 Quantum information theory and applied verification research06:36 Smart contracts, specifications, and provably correct systems07:11 If you create a super intelligent mathematician, then you solve most problems.‍Check out caj.al

A bug shows up in your product Lucent spots it Drops an alert in SlackSomeone tags a coding agent in the thread It ships a PRFixed before anyone knew it was brokenThis is the workflow Lucent users are actually running todayLucent watches your session replays like a human analyst: friction points, bugs, failed upgrades...Patterns no human has time to sit throughConnects to PostHog. Quick set up. Julius, Mastra, BrowserUse and more already on itEvery session replay you're collecting and not watching is data you're wastingLucent can help🎙️ Alisa Rae, Founder, Lucent on Fondo START pod 00:44 Building an AI that watches your session replays 1:10 Leaving Australia's Tall Poppy Syndrome for SF 2:41 Getting rejected by YC as a solo founder 3:32 Why she decided to keep building alone 4:36 Raising a $1.3M pre-seed round 6:43 Pivoting Lucent into AI training data 7:29 Returning to the original vision 9:27 Why this was impossible before modern AI 11:21 AI agents generating PRs from Lucent alerts 13:28 get a co-founder for YC?Check out www.lucenthq.com to learn more

The entire vibe coding wave got one thing backwardsEveryone optimized for speedPrompt in, app out, as fast as possibleBut the bottleneck isn't just speedIt's contextGive coding agents better context and everything improves.Lower costFaster executionBetter outputWoz is built for that shift.A Claude Code plugin that makes coding agents faster, cheaper, and higher performingUp to 55% cheaperUp to 40% fasterHigher benchmark scores than Claude Code aloneAnd the timing matters:The age of tokenmaxxing is ending. The age of more measurable ROI is beginning.🎙️ Ben Collins, CEO & Co-Founder, Woz on Fondo START pod 00:57 Context is the biggest bottleneck02:11 Why prompt-to-output breaks at enterprise scale03:28 Apple cracking down on AI-generated apps05:02 Why local AI models are becoming viable06:37 The rumored Cursor and xAI acquisition07:01 Coding agents and the zero-margin problem09:48 The end of token maxing12:32 Why AI infrastructure is shifting toward ROICheck out wozcode.com to learn more.

Your AI agent makes a mistake. Who pays for it?Not your E&O policyThat assumes a human made the errorNot your cyber policyThat assumes an attacker breached the systemAutonomous AI agents are explicitly excluded from bothInes Boutemadja discovered this while building AI agents for enterprisesThen procurement started asking for proof the AI was actually coveredExisting policies couldn’t provide it.So she built Klaimee: the first purpose-build E&O coverage for AI agentsBuilt for companies deploying AI in healthcare, fintech, and voice & moreHer bet on the future: AI liability is inevitable‍🎙️ Ines Boutemadja, CEO & Co-Founder, Klaimee on Fondo START pod ‍00:57 "Klaimee is the insurance for your AI agents" 02:11 Building therapeutic agents and insurance claims agents 02:48 Existing policies "completely excluded" agentic AI 03:31 Enterprise procurement blocking deals without coverage 05:02 Why cyber and E&O fail autonomous AI risk 06:01 AI liability coverage for agentic system providers 10:20 First Algerian woman founder in YC 11:15 "Don't wait for permission" 14:20 "AI liability is inevitable" 16:18 Why stealth mode is overratedLearn more at www.klaimee.ai‍

AI coding tools are making developers faster than everQA still hasn't caught upAakash saw it firsthand at Windsurf during the AI coding explosionThen realized:QA is probably a bigger bottleneck than coding itselfSo he built CanaryAn AI QA engineer that reads your source code, maps changes to real user flows, spins up remote browsers, and drops video recordings directly into the PR15 minutes instead of daysNow the workflows are getting even crazierOne customer already runs a fully autonomous loop:Linear ticket → Cursor background agent → PR → Canary QA → MergeThe engineer just reviews the outputs before approving the changeCode generation is becoming autonomousQA has to become autonomous too🎙️ Aakash Mahalingam, Co-founder & CEO, Canary on Fondo START pod ‍00:15 What Canary does 0:57 Spinning up remote browsers to test 1:42 User flow validation vs file reviews 2:29 How this changes developer workflows 3:12 Lessons from Windsurf and Cognition 3:43 QA is a bigger bottleneck than coding itself4:08 Pivoting inside YC from regression to feature testing 5:02 Applying to YC five times 6:11 Background agents opening PRs autonomously 6:37 "They didn't write a single line of code to merge that"‍Check out runcanary.ai to learn more. 

The “buy a list and blast emails” era is endingNot because outbound stopped workingBecause the data got smarter than the emailsKathryn Wu and her co-founder built Openmart as an SMB intelligence layer:Verified ownersDecision-maker contactsGoogle reviewsWebsite qualityLocation intelligence & moreNow they've launched "OpenClaw for Sales"A conversational outbound workspace built on top of that dataNo spreadsheet cleanupNo disconnected enrichment toolsNo giant table viewsJust ask the database what you needDoorDash uses Openmart for regional planningWhatnot uses it to identify high-quality sellersClay customers use the data directly inside outbound workflowsTry it out at openmart.com🎙️ Kathryn Wu, Co-Founder of Openmart on Fondo START pod‍01:13 Openmart’s SMB intelligence database and Openclaw for Sales02:16 The shift from giant lead lists to AI qualification and scoring03:43 Why SMB data differs from LinkedIn-centric sales platforms05:57 How modern GTM teams use multi-channel outbound workflows06:39 The highest-value qualification signals in SMB sales08:25 Openclaw for Sales and conversational outbound workflows09:02 How DoorDash uses Openmart for regional planning11:28 Deduplication and the hidden pain of outbound infrastructure12:16 “The future is qualification and scoring”14:08 Why proprietary data is the moat behind AI outboundLearn more at openmart.com

A drug was approved in ChinaThen it failed clinical trials in the USNo one could figure out whyThe answer was hiding in the gut microbiomeAnto Bio’s model helped identify it99% of the genes in your body aren't yours They belong to your microbiomeOver two-thirds of drugs are heavily affected by how your gut processes themBut the microbiome data was too noisy. Too messy. Petabytes of it just sitting thereAnto Bio's mission: make the gut microbiome computable for the first timeThey're building foundation models that help pharma design drugs that work for more people🎙️ Arvid Gollwitzer, Co-Founder, Anto Bio on Fondo START pod1:18 What Anto Bio is building and why it's novel2:10 Why this didn't exist before3:05 99% of your genes aren't yours4:00 Why pharma ignoring the microbiome is a huge problem5:04 The microbiome is modifiable6:30 Modeling microbiome differences across populations7:00 Drug approved in China, failed in the US, and how Anto found out why8:02 How many drugs are affected by microbiome metabolism10:00 GLP-1s and the gut microbiome connection12:42 The biggest unsolved variable in drug developmentLearn more at www.anto.bio

Most people think they need more motivation to work outThey need less frictionBut training plans assume your life stays staticYou travelIt rains You have 20 mins between meetingsPeople don't quit because they lack disciplineThey quit because adapting the plan is harder than the workout itselfHYBRD brain can helpAn adaptive AI agent that adjusts your training to whatever life throws at youBuilt by ultramarathoners, Ironman finishers, and a founding engineer who won a 100-mile cycling race🎙️ Matthew Ruiters, CTO & Co-Founder, HYBRD on Fondo START pod02:01 Connecting every wearable into one training system02:42 Using sports science to measure total training load03:07 How former Whoop teammates started building HYBRD03:50 Why “serious athlete” is really a mindset04:49 The idea behind HYBRD Brain05:18 Adapting workouts around travel, weather, and hotel gyms06:13 The consistency insight from 895,000 athletes10:44 Why athletes wanted adaptation, not just tracking11:35 Why friction stops founders from staying healthy14:03 How AI tools are changing software engineeringlearn more at www.hybrd.com

Every voice AI has the same problemThe moment you stop talking, it starts talkingBut humans don’t work that wayIn real conversations, pauses don’t always mean interruption. Sometimes they mean: keep goingCurrent AI misses this completely.Amit Yadav spent years building real-time AI systems at Meta Reality Labs for Ray-Ban smart glassesNow he’s building Fern: a sales copilot that assists teams live during meetings.Not after the call.During it.The hard problem isn’t retrieval anymore.It’s judgment.Because the future of AI won’t be defined by smarter answers alone.It’ll be defined by systems that understand how humans actually communicate.🎙️ Amit Yadav, CEO & Co-Founder of Fern, on Fondo START pod‍01:24 “Imagine your sales rep hears a prospect question and answers appear instantly.”01:55 The hardest AI problem: knowing the right time to help02:09 “Our in-house model beats human judgment, ChatGPT, and Gemini.”02:46 Building realtime AI systems for Ray-Ban Meta smart glasses03:57 Fern’s desktop copilot listens and assists during meetings04:34 The industry shift from prompted AI → proactive AI assistance05:28 Replacing Slack searches, docs, and tabs during live calls07:10 “The most exciting thing is AI helping you without asking for it.”11:09 Why silence in human conversation doesn’t always mean interruption14:15 The need for a realtime social-awareness layer between humans and LLMs‍Check out getfern.ai

Speed to lead sounds like a sales tacticIt's actually the whole gameRuben Harris knows this better than anyoneHe built a 300-person call center across 3countries to reach 1M people a month ...and saw exactly where the ceiling wasThe problem isn't effort. It's physics. A working-class student fills out a form at 10 PM. Nobody's staffed at 10 PMSo 90% of leads go coldNot because they weren't interestedBecause nobody got back to them fast enoughOutrival's AI agents respond in under a minuteThey call, text, email, and follow upOne client made $10M in new revenue from leads they already hadThe best outbound AI isn't replacing humansIt's doing the work humans never had time to do🎙️ Ruben Harris, Co-founder & CEO, Outrival on Fondo START pod‍01:12 Why Outrival focuses on outbound AI03:18 Building a 300-person call center05:27 Why speed-to-lead changes everything08:52 Aged leads as hidden revenue09:57 100,000 calls in an hour14:11 Turning one school’s leads into $10M revenue18:43 Enterprise sales advice for AI founders24:48 Pricing digital workers like employees27:56 From cello to Wall Street to Silicon Valley50:14 Building startups in the AI eraCheck out www.outrival.com"Turn every interaction into a long-term relationship"

The way to make software faster was always: make the code happen fasterAI solved that Cursor, Claude Code, Replit...Building is nearly instant nowSo what's slow?Knowing if it actually works.When AI writes your feature, someone still has to answer: did it break everything else?That's the new bottleneck. Not creation. ValidationJack Brown is building Lark to solve exactly this. An AI QA engineer that writes tests, keeps them updated, and continuously validatesThere's the building, and then there's the thing that lives on after the buildingLark is focused on that second half‍🎙️ Jack Brown, Cofounder, Lark on Fondo START pod00:15 Building an AI QA engineer01:56 AI dramatically increased development velocity02:15 The bottleneck becomes verification02:25 Cursor writes the code. Lark writes the tests.03:24 Acceptance criteria → automated validation04:15 Non-technical founders building software with AI05:35  Lark agents autonomously defining tests06:31 Everything is just validation07:56 Fully agentic software creation08:53 Why testing becomes more critical in AI-generated codebasesVisit getlark.ai to learn more

Google used to surface the PDF of a law if you searched its exact name.Now it surfaces consultancies advising on it.And LLMs have gotten worse at this too. As they’ve tried to cut costs and focus on the information 99.99% of users want, long-tail edge-case data has become much harder to find.Natalie Aresta-Katz was at a big law firm. One of their biggest clients wanted to track new and proposed laws in 45 countries that didn’t publish in English.They hired lawyers in every single country. Tried probably 15 tools on the market. Ran manual searches.Nothing worked well.Now she's building the tool she wished she had.One workflow went from up to 13 paralegals and around 120 hours a month to one user and about an hour and a half.‍🎙️ Natalie Aresta-Katz, CEO & Founder of Regbase, on Fondo START pod (full ep in comments)‍00:20 “Regbase was really born out of a place of necessity.”01:32 Tracking new and proposed laws across 45 countries02:00 Why existing tools failed02:43 The “Wizard of Oz” backroom behind regulatory databases03:26 Why long-tail legal data is getting harder to find04:23 Dropping into YC after Andrew said, “an MBA isn’t really school”05:37 Finding unspent state-level funding for school safety upgrades07:05 From 120 hours to about 1.5 hours07:38 How Regbase uses LLMs, specialty search, and link-following08:33 Why legal work may shift from hourly billing to output09:38 Searching laws in languages teams don’t speak10:27 Why Regbase unlocks more opportunity for law firms11:00 Who should book a demo: government vendors, corporate lawyers, AI law trackers, employment lawyers, and teams tracking high-volume regulatory changevisit regbase.com to learn more