Skip to main content
AmigoFish

20VC with Harry Stebbings

20VC with Harry StebbingsVideoLast show:

Not rated yet
CosmosOpen at the sourceReport a problem

About this show

20VC, hosted by Harry Stebbings, takes you inside the world of Venture Capital, Startup Funding and The Pitch. Join Harry and discover how you can attain funding for your business by listening to what the most prominent investors are directly looking for in startups, providing easily actionable tips and tricks that can be put in place to increase your chances of getting funded. The Twenty Minute VC also provides an instructional guide as to what it takes to get employed in the Venture Capital industry, with VCs giving specific advice on how to get noticed from the crowd and increasing your chances of employment. Our amazing Venture Capitalists also provide their analysis of the current technology market, providing advice and suggestions on the latest investing trends and predictions. Join us so you can see how you can get big, powerful improvements, fast.

Videos come from the channel's public YouTube feed. The artwork is the newest video thumbnail. Last updated . Names, artwork and text belong to their publishers.

Cosmos

Similar by topic and creator

See the full Cosmos

Videos

  • Jev Creator on Will US Open Models Catch Chinese | The Cost of Intelligence & Tokens | Diogo Almeida

    · 1 hr 13 min

    Diogo Almeida is the Co-Founder and CEO of TypeSafe AI, the company behind Jev. Jev is an AI model that lets software make decisions automatically and tells developers how confident it is in each call. According to Sequoia's Pat Grady, it went from zero to $100M in revenue in seven days. TypeSafe's latest round valued the company at $7.5BN, with Martin Casado at Andreessen Horowitz leading and Sequoia also investing. Before TypeSafe, Diogo was a researcher at OpenAI, where he helped invent ChatGPT. ----------------------------------------------- Timestamps: 00:00 Intro 01:43 Why AI Reliability Is the Biggest Bottleneck to Automation 04:51 Why Data Matters More Than Compute 07:30 Why Scaling Compute Alone Stopped Working 09:05 What Actually Makes AI Training Data Valuable 14:00 Is Synthetic Data Overhyped? 15:06 Why Diogo Thinks Most Neo Labs Are “Incredibly Wasteful” 17:18 What Everyone Gets Wrong About Jev 19:46 Why OpenAI Can’t Simply Copy Jev 20:28 Why Benchmarks Don’t Tell You Which Models Are Best 23:02 Why Decision Models Could Dwarf LLM Token Usage 28:12 Why Enterprise Software Has Barely Changed Despite AI 29:32 Do Incumbent SaaS Companies Have the Advantage in AI? 30:57 Why Diogo Disagrees That Every Company Should Own Its Own Intelligence 33:34 Are Specialized Enterprise Models Actually Worth It? 37:00 Is Harvey Too Thin a Wrapper on Frontier Models? 39:44 Is Intelligence Per Token Really Improving? 42:19 Why AI Benchmarks Are “Total Bullshit” 45:44 Will US Open-Source Models Catch Chinese Models? 48:34 Why Diogo Thinks the Answer Is Probably “No” 49:52 Why Diogo Thinks AI Is Already in a Bubble 53:16 How Jevons Paradox Could Create Massive AI Demand 57:06 Inside TypeSafe’s $750M Raise at a $7.5BN Valuation 1:00:49 Why Diogo Chose Andreessen and Sequoia 1:04:43 Why TypeSafe Refuses to Pay $20M AI Compensation Packages 1:09:22 Why Diogo Doesn’t Hire OpenAI Researchers 1:10:50 Why Robotics Is the Most Overhyped AI Category 1:11:17 The Most Underhyped Opportunity in AI ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZTtgTNBKwtZBMHvl?si=85bc9196860e4466 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-twenty-minute-vc-20vc-venture-capital-startup/id958230465 Follow Harry Stebbings on X: https://twitter.com/HarryStebbings Follow Diogo Almeida on X: https://twitter.com/CompleteSkeptic Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/contact ----------------------------------------------- #20vc #harrystebbings #diogoalmeida #typesafeai #jev #ai #openmodels

    Not rated yet
  • The AI Compute Crisis: Why We Need Thousands of Times More Computing Power | Fractile Founder

    · 1 hr 15 min

    Walter Goodwin is the founder and CEO of Fractile, the British chipmaker taking on Nvidia in the race to make AI models answer faster. Fractile has raised $940M, including a $720M Series C that valued the company at $7.5BN. Thrive, Founders Fund, Lightspeed and Redpoint led the latest round. ----------------------------------------------- Timestamps: 00:00 Intro 02:03 Can a Chip Company Win Without Going Full Stack? 04:04 Why Every AI Company Is Fighting Commoditization 05:13 What Parts of the AI Stack Are Actually Commoditized? 09:16 Why Specialized AI Chips Could Reshape the Market 11:10 Can Specialized Chips Take Share From Nvidia? 14:08 Why Memory Bandwidth Is the Real Bottleneck in AI Inference 18:50 How 3D Memory Stacking Could Transform AI Chips 23:52 Why Long-Running Agents Require a New Kind of Chip 27:06 Why Most AI Chip Startups Struggle to Reach Real Production Scale 29:12 Why Fractal Chose Not to Use Broadcom 35:39 Why Data Center Capacity Is Becoming a Bigger Bottleneck Than Silicon 39:50 Is China Structurally Ahead of the West on AI Energy? 42:26 Why Inference Demand Could Be Far Bigger Than Anyone Expects 48:21 How Do We Fund the Enormous Compute Requirements of AI? 51:12 What Happens to the Cost of Inference Over the Next Decade? 55:11 Which Parts of the AI Stack Will Capture the Least Value? 1:00:30 Why Nvidia Bought Groq 1:02:52 Modal vs Fireworks vs Baseten: Who Wins Inference? 1:04:07 Will Inference Produce $100BN Companies? 1:04:48 Fractal Raises a $720M Series C 1:09:06 Why Walter Would Buy TSMC 1:11:28 Why AI Capability Timelines Are Compressing Faster Than Expected 1:12:38 Will Nvidia Be a $10TRN Company? 1:14:13 What the UK Must Do to Become an AI Superpower ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZTtgTNBKwtZBMHvl?si=85bc9196860e4466 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-twenty-minute-vc-20vc-venture-capital-startup/id958230465 Follow Harry Stebbings on X: https://twitter.com/HarryStebbings Follow Walter Goodwin on X: https://twitter.com/goodwin_ml Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/contact ----------------------------------------------- #20vc #harrystebbings #founder #ceo #nvidia

    Not rated yet
  • Sequoia Leaves London, Close Scout Program & End Evergreen Fund | NASDAQ Loses $2.5T in 24 hours

    · 1 hr 7 min

    Jason Lemkin is one of the leading SaaS investors of the last decade with a portfolio including the likes of Algolia, Talkdesk, Owner, RevenueCat, Saleloft and more. Rory O’Driscoll is a General Partner @ Scale where he has led investments in category leaders such as Bill.com (BILL), Box (BOX), DocuSign (DOCU), and WalkMe (WKME), among others. ----------------------------------------------- Timestamps: 00:00 Intro 01:01 Sequoia Restructures: London, Scouts and Evergreen 01:41 Why Sequoia Is Ending the Evergreen Fund 07:05 Why Sequoia Thinks It Can Cover Europe From the US 09:13 Why Sequoia Is Ending the Scout Program 11:56 Why Founders Fund’s Concentration Strategy Works 12:22 Jensen Huang Investigation: What Happens Now? 14:19 Why Jason Thinks Jensen “Needs a Pass” 17:30 Why China Could Be the Biggest Loser From the Nvidia Investigation 20:08 NASDAQ Loses $2.5TRN: Is the AI Trade Cracking? 22:29 Why Anthropic’s IPO Matters More Than the AI Selloff 23:28 Why Anthropic Is Pricing Its IPO Conservatively 25:27 Will Anthropic’s IPO Pop or Trade Flat? 26:12 Is There Enough Capital to Keep Funding Anthropic? 27:17 Why Anthropic Is in a Better Position Than OpenAI 29:09 OpenAI Reaccelerates: Enterprise Becomes the Majority of Revenue 30:01 What Happens When AI Token Prices Keep Falling? 31:29 Can Persistent Consumer Agents Ever Have Good Unit Economics? 34:17 Investment Committee: Zipline vs fal vs Revolut 37:05 Why Revolut Could Become Europe’s Biggest Bank 41:16 Why AI API Keys Are Becoming One of Cybercrime’s Biggest Targets 45:05 Why Security M&A Is Accelerating 47:29 Why Older SaaS Companies Are Struggling to Become AI-Native 51:48 Why Cloudflare Is Winning From the Agent Boom 54:31 Why Seat-Based SaaS Is Under Pressure From Agents 55:21 Figma Below $10BN: Did It Miss Its Best Exit? 59:37 What Figma Needs to Do to Win in Agentic Design 1:02:16 Why California Rejected the Billionaire Wealth Tax 1:04:28 Why Losing Founders From California Would Hurt the AI Ecosystem ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZ... Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast... Follow Harry Stebbings on X: https://x.com/harrystebbings Follow Jason Lemkin on X: https://x.com/jasonlk Follow Rory O’Driscoll on X: https://x.com/rodriscoll Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/con... ----------------------------------------------- Legal Disclaimer: The content of this podcast is for informational and entertainment purposes only and does not constitute financial or investment advice. Any discussion of stocks, public markets, or investment strategies reflects the personal opinions of the speakers and should not be relied upon when making investment decisions. Figures, valuations, and financial data referenced may be estimates or subject to error. Always consult a qualified financial adviser before making any investment decision. The views expressed are those of the individual speakers and do not represent the views of 20VC or its affiliates. ----------------------------------------------- #20vc #harrystebbings #roryodriscoll #jasonlemkin #jensenhuang #anthropic #sequoia #nvidia #ai

    Not rated yet
  • WE SPEND MORE ON AI THAN SALARIES | MERCOR CEO

    · 49 sec

    Mercor Co-founder & CEO, Brendan Foody, on the company's high token spend...it's more than their salary spend!

    Not rated yet
  • HE COULD WALK AWAY WITH BILLIONS. HE SAID NO

    · 32 sec

    Brendan Foody (Co-founder & CEO, Mercor) on why building a legendary company matters more than cashing out

    Not rated yet
  • Why WashU’s CIO Is Still Investing in China

    · 36 sec

    Scott Wilson, Chief Investment Officer at Washington University in St. Louis, explains why his team continues to invest in China and how they weigh geopolitical risk against opportunity.

    Not rated yet
  • Mistral CEO on Why AI Inference Is Becoming a Commodity

    · 57 sec

    Mistral CEO Arthur Mensch explains the two sides of AI inference, from on-demand pricing to reserved capacity with guaranteed speed and reliability.

    Not rated yet
  • Mistral CEO Explains Why AI Models Are Becoming Defence Systems

    · 38 sec

    AI models are becoming so powerful that they're increasingly being treated as defence systems. Arthur Mensch, Co-Founder and CEO of Mistral AI, explains why the US and China are tightening controls on AI technology, and why securing access to models is becoming just as important as securing compute.

    Not rated yet
  • Mistral CEO: Why OpenAI & Anthropic Will Come for Your Business | Are Chinese Models Truly Safe?

    · 50 min

    Arthur Mensch is the Co-Founder and CEO @ Mistral, the last remaining hope for frontier intelligence in Europe. Mistral has raised approximately $7.5BN, including debt, and reached a valuation above $24BN. Previously a researcher at Google DeepMind, Arthur co-authored the landmark Chinchilla research, showing how smaller, better-trained AI models could outperform much larger rivals. ----------------------------------------------- Timestamps: 00:00 Intro 01:07 Why Europe Needs Its Own Frontier AI Companies 02:13 Why Sovereign AI Is an Economic Necessity, Not Just Politics 04:01 What “Sovereign AI” Actually Means for Enterprises 06:27 Are Chinese Open-Weight Models Really Safe? 09:18 Why Critical Industries Are Reluctant to Use Chinese Models 11:17 Why OpenAI and Anthropic’s Closed Models Create Too Much Centralized Power 13:43 Why Energy Is Not the Biggest AI Bottleneck Today 16:42 What Mistral Learned About Buying and Forecasting Compute 18:47 Why Europe Still Does Not Have Enough Compute 20:03 What European Governments Need to Do to Accelerate AI Infrastructure 22:17 Does Every Country Need Its Own Sovereign AI Model? 24:18 Why Mistral Is Going Full Stack—but Not Building Its Own Chips 28:12 What Mistral’s Business Model Could Look Like in Five Years 31:27 Why the Inference Market Behaves Like a Commodity Market 33:00 What Separates Winning AI Labs From the Rest 34:39 Why Compute Access Is the Biggest Bottleneck for New AI Labs 37:26 Are $100BN AI Training Runs Actually Coming? 38:24 Why Public Markets Are Punishing AI Infrastructure Companies 39:47 Why Mistral Avoided the Consumer AI Market 41:00 Why Self-Improving AI Still Needs Well-Defined Problems 42:32 Does Mistral Feel the Pressure of Being Europe’s Best Shot at Frontier AI? 44:17 Can AI Add More Than 2% to GDP Growth? 46:04 Why Sovereign AI Could Become the Most Important Enterprise AI Category 46:35 How AI Will Change Every Engineering Profession 47:28 Why Mistral Could Eventually IPO in Europe 48:31 Why AI-Driven Scientific Discovery Excites Arthur Most ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZTtgTNBKwtZBMHvl?si=85bc9196860e4466 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-twenty-minute-vc-20vc-venture-capital-startup/id958230465 Follow Harry Stebbings on X: https://twitter.com/HarryStebbings Follow Arthur Mensch on X: https://twitter.com/arthurmensch Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/contact ----------------------------------------------- #20vc #harrystebbings #mistral #ai #inference #openmodels

    Not rated yet
  • Inside WashU's Endowment Fund | SpaceX Made Up 20% of Out AUM | Scott Wilson

    · 57 min

    Scott Wilson is the CIO of WashU, one of the best-performing endowment funds in the world. Under his leadership, Scott has grown the endowment from $8BN to $19BN today. At one point, SpaceX was over 20% of the entire value of the endowment. ----------------------------------------------- Timestamps: 00:00 Intro 02:28 Why Scott Redeemed Over Half the Portfolio in His First Week 04:13 How WashU Grew From ~$8BN to Nearly $20BN 06:16 Why Scott Would Put 70–80% in Privates Without Payout Constraints 07:56 Why WashU Puts More Than Half Its Private Book Into Directs and Co-Invests 09:17 Why Direct Investments Have Driven WashU’s Outperformance 11:52 Why Fund Concentration Limits Hurt Venture Returns 15:00 Scott’s Biggest Investing Mistake: Not Adding Enough to Winners 19:00 Why Scott Still Avoids Mega-Funds 24:56 Why Scott Thinks Private AI Valuations Are Showing “Stupidity” 27:51 Why Scott Thinks Many AI Companies Won’t Exist in 10 Years 29:02 Why Frontier Models May Struggle to Sustain Pricing Power 32:21 Why WashU Walked Away From Sequoia’s Evergreen Structure 35:29 The #1 Reason WashU Drops a Venture Manager 40:34 Why Scott Expects an AI Shakeout 43:05 How WashU Generated ~$17BN of Profit in Nine Years 44:39 The Mistake That Cost WashU Billions 45:55 Why Scott Regrets Not Selling More in 2021 47:35 Why Scott Still Likes Investing in Europe 49:57 Why WashU Is Still Investing in China 54:09 Why Scott Thinks Lee Fixel Is One of the Best Investors He Knows ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZTtgTNBKwtZBMHvl?si=85bc9196860e4466 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-twenty-minute-vc-20vc-venture-capital-startup/id958230465 Follow Harry Stebbings on X: https://twitter.com/HarryStebbings Follow WashU on X: https://twitter.com/WashU Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/contact ----------------------------------------------- #20vc #harrystebbings #investing

    Not rated yet
  • Who's REALLY at risk in the AI boom?

    · 25 sec

    Rory O'Driscoll (Scale Venture Partners) on where the risk in AI is right now

    Not rated yet
  • Google Loses $500B in 12 Hours | DoorDash Buys Instacart for $17B | Kalshi Raises $1B

    · 1 hr 8 min

    Jason Lemkin is one of the leading SaaS investors of the last decade with a portfolio including the likes of Algolia, Talkdesk, Owner, RevenueCat, Saleloft and more. Rory O’Driscoll is a General Partner @ Scale where he has led investments in category leaders such as Bill.com (BILL), Box (BOX), DocuSign (DOCU), and WalkMe (WKME), among others. ----------------------------------------------- Timestamps: 00:00 Intro 00:54 OpenAI Cuts Its Valuation to $1.25TRN 02:02 Is the AI Trade Starting to Crack? 04:58 Alphabet Loses ~$500BN 07:08 Why Anthropic’s IPO Could Face Pricing Pressure 08:02 Is Anthropic’s Growth Actually Slowing? 11:19 Why AI Model Pricing Is Falling Faster Than Expected 13:50 Why AI Economics Are Becoming Almost Impossible to Predict 16:11 Is Canva Heading Toward Quiet Obsolescence? 19:44 Can Consumer AI Agents Ever Have Sustainable Unit Economics? 26:36 DoorDash Buys Instacart for $17BN 27:52 Are AI Agents an Existential Risk to DoorDash? 31:56 Kalshi Raises $1BN at a $40BN Valuation 32:28 Cohere’s Sovereign AI Bet With the Canadian Government 38:35 How AI Pen-Testing Tools Could Become Cyber Weapons 41:10 Cognition vs Baseten vs Serval 44:40 Why Serval Could Be the Best Risk-Reward of the Three 47:54 Can Vertical AI Actually Expand the TAM? 50:00 Why Inference Still Has the Fastest Revenue Velocity in AI 52:24 Can Crosby Reinvent the Law Firm With AI? 59:02 Physical Intelligence at $20BN: How Big Can Robotics Get? 1:02:00 Why Venture Needs M&A to Keep Working 1:03:55 Why Salesforce Is Buying Its Way Into the AI Era 1:07:04 Why Big Tech Acquirers Want $100M+ Revenue Before They Buy ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZ... Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast... Follow Harry Stebbings on X: https://x.com/harrystebbings Follow Jason Lemkin on X: https://x.com/jasonlk Follow Rory O’Driscoll on X: https://x.com/rodriscoll Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/con... ----------------------------------------------- Legal Disclaimer: The content of this podcast is for informational and entertainment purposes only and does not constitute financial or investment advice. Any discussion of stocks, public markets, or investment strategies reflects the personal opinions of the speakers and should not be relied upon when making investment decisions. Figures, valuations, and financial data referenced may be estimates or subject to error. Always consult a qualified financial adviser before making any investment decision. The views expressed are those of the individual speakers and do not represent the views of 20VC or its affiliates. ----------------------------------------------- #20vc #harrystebbings #roryodriscoll #jasonlemkin #openai #cognition #alphabet #google #doordash #ai

    Not rated yet
  • Why He Waited Until 37 to Build a $15.7BN Company

    · 29 sec

    What held Erik Bernhardsson back from starting a company? Modal Labs founder explains how a lack of confidence delayed his journey into entrepreneurship, and why he now wishes he'd started earlier. Five years later, Modal is valued at $15.7BN.

    Not rated yet
  • The $15B Founder Who's Never Made a Pitch Deck

    · 20 sec

    Modal founder Erik Bernhardsson explains why building investor relationships early is the key to fundraising

    Not rated yet
  • Great brands are built by being different

    · 24 sec

    Everyone wants a standout brand Few are willing to stand out Erik Bernhardsson, Co-Founder & CEO of Modal, on Diogo Almeida's unconventional approach to building Jev's brand and community

    Not rated yet
  • CEO of Modal on AI Inference

    · 58 sec

    Erik Bernhardsson, Co-founder and CEO of Modal on 20VC 🎧️

    Not rated yet
  • Will AI Replace Software Engineers?

    · 50 sec

    Will AI replace software engineers? Erik Bernhardsson, Co-founder and CEO of Modal, explains why AI could actually increase demand for developers rather than replace them.

    Not rated yet
  • Modal CEO: Has Open-Source LLM Inference Peaked? | What Happens to the Cost of Compute

    · 1 hr 2 min

    Erik Bernhardsson is the Co-Founder and CEO of Modal, the AI infrastructure platform for training models, running inference and powering AI agents. Modal has raised over $1.2 billion, including a recent $750 million Series D led by Accel at a $15.75BN valuation. The company has surpassed $1BN in ARR, growing more than 15x year-on-year. Previously, Erik built Spotify's original music recommendation system. ----------------------------------------------- Timestamps: 00:00 Intro 01:42 How Erik Fell in Love With Coding at Eight Years Old 03:43 Why Erik Rejects the “Eating Glass” Founder Narrative 05:32 Why Erik Hates Fundraising 05:56 Why Erik Has Never Made a Pitch Deck 06:25 Why Founders Shouldn’t Always Maximise Valuation 08:03 Why Early-Stage Startups Don’t Really Have Moats 08:52 Why Brand Matters More Than Ever in AI Infrastructure 11:43 Why Inference Is Becoming the New Cloud 12:32 Why Pure Inference Becomes a Commodity Business 13:25 Where the Gross Margins in Inference Will Come From 15:00 Why the Inference Market Will Fragment Into Specialized Products 16:33 Inside the “Open-Air Drug Market” for GPUs 17:30 Why Modal Wants to Aggregate the World’s Compute 19:05 How Modal Forecasts GPU Demand Growing 34% Month-on-Month 20:31 What Happens to GPU Prices Over Time? 21:14 Why AI Compute Demand Is Still Extremely Early 22:43 Does Specialized AI Hardware Threaten Nvidia? 23:41 Why CUDA Is Not Nvidia’s Real Moat 24:38 Can Neo-Clouds Successfully Move Up the Stack Into Inference? 26:47 What Happens to Open Models as Frontier Models Get Cheaper? 28:01 Can Europe Produce a Competitive Open Model? 28:39 Why Nvidia Is Backing the US Open-Model Ecosystem 29:52 Why Specialized Models Can Become Huge Businesses 30:25 The Bull Case for Jev at $100BN 32:57 How Modal Plans to Expand From a Product Into a Platform 34:41 How Modal’s Sandbox Business Reached a $300M Run Rate 35:46 Why Infrastructure Products Can Take Years to Find Their Killer Use Case 37:41 Why Modal Underinvested in Open-Model Inference 38:19 Has Open-Model Inference Already Peaked? 39:11 Modal Hits a $1BN Revenue Run Rate 39:38 How Big Can Modal Be in 12 Months? 40:55 Modal’s Next Big Revenue Opportunities 42:28 How Important Will Post-Training Become? 44:13 Why AI Agents Won’t Kill Software Engineers 45:26 Modal Spends ~$500K Per Month on Coding Agents 46:06 Will Coding-Agent Spend Keep Growing as Models Get Cheaper? 47:11 Why AI Training Data Is Such a Lucrative Market Today 48:46 Should Every Company Own Its Own Intelligence? 51:24 Is Model Routing a Defensible Business? 52:40 Why Erik Would Have Told OpenRouter Not to Sell 53:45 Why Erik Expects a 20–30% AI Market Pullback 54:18 Why Employee Secondaries Matter in the AI Talent War 55:25 Why Top AI Labs Are Paying Exceptional New Grads $20M+ 57:10 Why Erik Would Back TurboPuffer 57:39 What Erik Changed His Mind on About the Sandbox Market 58:04 Which AI Companies Erik Would Avoid Investing In 1:01:07 Why AI Drug Discovery Excites Erik Most ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZTtgTNBKwtZBMHvl?si=85bc9196860e4466 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-twenty-minute-vc-20vc-venture-capital-startup/id958230465 Follow Harry Stebbings on X: https://twitter.com/HarryStebbings Follow Erik Bernhardsson on X: https://twitter.com/bernhardsson Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/contact ----------------------------------------------- #20vc #harrystebbings #erikbernhardsson #ceo #modal #ai #opensource #inference

    Not rated yet
  • David Tisch: Why No Manager Wants to Stay Small & Why Competing with Mega Platforms is a Losing Game

    · 1 hr 20 min

    David Tisch is the Founder of Box Group, one of the best seed funds of the last decade with early investments in Cursor, Baseten and Clay to name a few. Today the fund invests from two funds, a $275M seed fund and a $275M opportunity fund. ----------------------------------------------- Timestamps: 00:00 Intro 01:25 Has Venture Capital Fundamentally Changed? 02:49 Why This Market Is Nothing Like 2021 05:59 Are $10–20M Seed Rounds the New Normal? 07:11 Why Bigger Outcomes Make Bigger Seed Rounds Rational 08:00 Why Lower Dilution Changes Venture Fund Math 10:21 Why Ownership Matters Less Than Getting Into the Outlier 11:48 Why Multi-Stage Firms Have Become So Powerful at Seed 14:13 Why Concentrated Seed Funds Overstate Their Win Rates 16:28 Does Too Much Capital Too Early Hurt Founders? 17:15 Why David Thinks the Lean Startup Was “Total Fake” 19:19 Why Founders Have All the Leverage in Today’s Market 22:03 How BoxGroup Invests Without Ownership Requirements 23:55 Why Getting to Deals Late Is the Biggest Reason VCs Lose 26:06 Are Lead-Only Seed Funds Becoming Obsolete? 28:54 Why Pure-Play Seed Is One of the Hardest Venture Strategies Today 32:27 How Cursor Returned the Fund Multiple Times 33:30 Why Company Creation Is Moving Faster Than Ever 35:21 Are the Best Companies Obvious From Day One? 39:00 Why Today’s Winning Founders Are More Technical—and Better 41:53 Why BoxGroup Stopped Investing in Consumer Brands at Seed 44:00 How Rogo Went From 39 Passes to a Consensus AI Winner 48:24 Are Pro Rata Rights Actually Worth Anything? 52:30 When Should VCs Take Money Off the Table? 57:29 Why Venture Is Consolidating Around Fewer Winning Firms 1:02:23 Finding vs Picking vs Winning: What Actually Makes a Great VC? 1:05:57 Why BoxGroup Doesn’t Allow “Grenades” in Investment Committee 1:09:56 The Best Seed, Series A and Growth Funds in Venture 1:13:06 Why the Best Venture Firms Will Keep Compounding 1:16:43 What David Is Most Excited About for the Next Decade ---------------------------------------------------------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZ... Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast... Follow Harry Stebbings on X: https://x.com/harrystebbings Follow Jason Lemkin on X: https://x.com/jasonlk Follow Rory O’Driscoll on X: https://x.com/rodriscoll Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/con... ----------------------------------------------- #20vc #harrystebbings #venturecapital #investing

    Not rated yet
  • OpenAI at $1.25 Trillion Is a Bargain?

    · 49 sec

    Jason Lemkin believes OpenAI is one of the most misunderstood investment opportunities in AI. At a $1.25 trillion valuation, he'd put 20% of a fund into OpenAI, arguing that the investment alone could return the entire fund. Is this a rare market opportunity or a massive gamble?

    Not rated yet