Hi {{first_name}}!
Remember when starting a company meant raising, hiring dozens of engineers, and hoping to find product-market fit before running out of runway?
Those days may be officially behind us.
In 2025, solo founders are building $80M exits in six months, and we’re experiencing the biggest computing revolution ever. Even Tesla's former AI chief is calling it "Software 3.0."
As we continue to build at Context Ventures, we've launched Startup Intros with a simple mission: to help early-stage founders find the right investors, faster and smarter.
But first, you should check out our events this week:
🔥 Events Coming Up
June 30: WorkTech Startup Pitch Night: Join us for a high-energy event where founders and investors come together to explore how AI is transforming the future of work. Unfortunately, pitches are closed but join on Luma!
July 2: Caffeine & Capital: The next round of our morning coffee series. Come ready to get caffeinated and talk startups. Sign up on Luma!
July 7: Live AdvancingVC Podcast Interview Series with Tim Hsia from Context VC:
10:30 AM: Amanda Robson, Founder of Modern Technical Fund, which recently closed a $22M debut fund. Sign up on Luma!
2:00 PM: Chris Douvos, Founder of Ahoy Capital, who is known for his unique LP perspective. Sign up on Luma!
3:30 PM: Jeff Clavier, the Founding Partner at Uncork Capital, one of the valley’s top seed-stage VCs. Sign up on Luma!
Keep reading for the latest edition of the Weekly Download covering the past week:
🎯 BLUF: Bottom Line Upfront
Computing Revolution: Vibe programming is making software development accessible to everyone
AI Speed Records: AI startups hitting $100M revenue in single-digit months, grabbing nearly half of all VC money
Mental Reality: Success requires thinking and executing 10x faster than traditional business school taught us
Funding Disruption: Companies now skip early fundraising and go straight to profitability first
Winning Strategy: Human-AI teamwork with user control beats letting AI run wild
Infrastructure Race: Smart companies are rebuilding their systems for AI customers
🧠 We're Living Through Computing's Third Big Change
Andrej Karpathy: Software Is Changing (Again) - YC AI Startup School
The numbers tell a crazy story. AI startups grabbed 46.4% of all US VC funding in 2024, that's $209B total, compared to less than 10% just ten years ago. But this isn't just about money; it's about a totally different way of building software.
Andrej Karpathy, Tesla's former AI chief who helped create OpenAI, calls this "Software 3.0." For 70 years, we told PCs what to do with code. Now we just talk to them in English.
Here's how we got here:
Software 1.0: Writing code or telling computers step-by-step what to do
Software 2.0: Training neural networks or teaching computers with data
Software 3.0: Talking to AI in English by prompting/vibe coding (we are here)
Think of it like the 1960s computer era all over again, AI compute is expensive and lives in the cloud, so we're all sharing time with these massive "AI computers" run by companies like OpenAI and Google.
But here's the game-changer: you don't need a CS degree to use them. You just need to know how to describe what you want.
The proof is everywhere: 25% of Y Combinator's newest startups write 95% of their code using AI, letting tiny teams hit $10M in revenue with fewer than 10 people.
🤯 The AI Speed Records That Are Breaking Everyone's Brain
The Impact of Vibe-Coding & AI on Venture Capital as an Asset Class - Data Driven VC
The growth numbers are absolutely wild. Stripe says AI startups are growing faster than SaaS companies ever did, with:
Cursor, the coding assistant, hit over $100M revenue
Lovable reaching $17M in recurring revenue in just 3 months, and
Bolt getting to $20M in 2 months flat.
This is happening everywhere:
Y Combinator's newest batch hit 10% weekly growth, "something that's never happened before in early-stage venture"
Last week, Base44's solo founder got an $80M exit to Wix in just 6 months
AI startups get billion-dollar evaluations in 2 years vs for typical SaaS in 9 years
Character.AI got to a $2.5B Google buyout in 24 months with just 30 employees
🎪 How VCs Are Freaking Out (And Adapting)
The whole VC world has been turned upside down. The old funding playbook is basically dead as companies skip the traditional fundraising and go straight to making money:
Traditional Fundraising: Pre-Seed (Pre-Revenue) → Seed → Series A
New Model: Bootstrap → Make Money → Maybe Raise (or Just Exit).
Here's what's happening:
Deal volume actually dropped even though total funding exploded.
79 deals over $100M in Q1 2025 vs 90 the quarter before, but total investment hit $80.1B because of massive rounds.
Translation: VCs are getting pickier, but when they write checks, they're writing big ones.
Here are the new VC stress points:
Speed panic: Companies go from idea to exit faster than VCs can do their homework
Signal vs Noise: Finding the good stuff among 13,700 new startups launching every day
Value-add Pressure: Just having money isn't special anymore; founders need real help with customers and growth
The winning VCs will help founders get customers, build distribution, and move faster.
As one NEA partner put it: "Companies will move past simple 'ChatGPT wrappers' to build AI workers that can actually think and take action to make real business impact."
🎮 The Winning Formula Part 1: Human-AI Advantage
Andrej Karpathy: Software Is Changing (Again) - YC AI Startup School
Here's where the strategy gets interesting. The winners aren't building fully automated AI robots, they're building AI tools that work with humans. And they are Iron Man's suit, not a robot that replaces Iron Man.
Take Cursor (which is quickly stealing VS Code users):
Low Help: Just auto-complete your typing
Medium Help: Rewrite entire functions for you
High Help: Make changes across your whole codebase
Human-AI teamwork works because AI is like having a really smart person with some weird blind spots. The winners will let users control how much AI help they want. For example, I love Perplexity, because it lets you pick how deep you want your search.
Forward-thinking companies are rebuilding their systems to act human-like by:
Build for commands, not clicks: AI can run code but can't click buttons
Write docs for robots: Use simple text instead of "click the blue button" tutorials
Create AI-friendly entry points: Think of files that tell AI exactly what your service does
Real example: Vercel gives you copy-paste terminal commands instead of telling you to "click here, then click there" because AI is amazing at running commands but terrible at navigating websites.
⚡ The Winning Formula Part 2: Speed Meets Smarts
The Impact of Vibe-Coding & AI on Venture Capital as an Asset Class - Data Driven VC
But here's the thing, NFX's James Currier figured out:
The biggest roadblock isn't the tech, it's our own thinking.
AI demands something totally different as most founders are still moving at "school speed" or "corporate speed."
The new gold standard: 20-100 experiments per week.
Currier puts it perfectly: "AI won't worry about what college friends think. It will just ship the product. AI won't procrastinate. For humans to keep up with AI, they need to think clearly and move fearlessly."
The mental shifts that separate winners from wannabes:
Your product isn't you: Ship something imperfect instead of trying to make it perfect
Copy shamelessly: Facebook copied Friendster, Google copied everything, it's fine (check out same.new)
Test everything: Treat features like experiments, not final products
What does this mean for you?
Starting a business has never been easier, but you need to move way faster than before.
If you're not testing 20-100 ideas every week, you're already behind.
Leverage AI and vibe coding tools to turn months of work into days.
What are you building at AI speed?
💸 What Other Startups are Popping Off
Here’s a roundup of trending startup activity this week:
🚀 AI No-Code Dev Platforms
Bolt.new: Currently hosting the World’s Largest Hackathon, where you can no-code to build with AI for a shot at some of the $1M+ in prizes
Warp 2.0: Groundbreaking agentic development environment that transforms coding into an intelligent, collaborative experience
Pythagora 2.0: All-in-one AI dev platform that revolutionizes how you build software from concept to deployment
Softr Databases: Powerful database builder that lets you create and manage data structures
HeyBoss AI Boss Mode: Complete AI workforce that builds your website and runs your entire business
🤖 AI Agents & Systems
AgentX 2.0: Ultimate cross-vendor multi-agent platform where different AI models collaborate
SmythOS: Revolutionary open-source agent operating system that orchestrates AI agents at scale
Zen Agents (by Zencoder): Collaborative AI agent builder with org-wide sharing and massive tool integration
🎨 AI Productivity & Content Tools
Tila AI: Mind-blowing infinite canvas that lets you create, code, search, and design AI content
FlashDocs API: AI-powered API that generates custom Slides and PowerPoint presentations automatically
ComputerX: Your brilliant AI agent that handles all your computer work
Runbear: The perfect AI team member that integrates seamlessly into Slack
NativeMind: Completely private, open-source AI assistant that runs entirely on your device for privacy and control
🛠️ Developer Tools & Infrastructure
Liveblocks 3.0: Ready-made AI copilots and real-time collaboration features that make your product more engaging
Ops AI by Middleware: Intelligent observability copilot that resolves production issues
AI Assistant by Mintlify: Conversational AI agent that transforms your documentation into an interactive experience
📱 CRM & Relationship Management
Twenty: The definitive open-source CRM that's redefining customer relationship management
Pally - AI Relationship Management: Intelligent connection manager that unifies all your social relationships
🎙️ Voice & Communication AI
🌟 Editor's Note
At Startup Intros, our mission is to keep you informed and help you raise in the fast-paced world of Silicon Valley. Our free weekly newsletter delivers curated insights, deals, and trends to help you navigate the startup ecosystem.
Stay tuned for our upcoming paid daily newsletter and fundraising app to help you close your next round!
💭 Parting Thoughts
Karpathy thinks the next 10 years will be about gradually giving AI more control, starting with AI assistants and eventually AI coworkers. But right now, the companies winning are the ones that let humans stay in charge while AI does the heavy lifting.
The bottom line:
How fast you can test and learn matters more than having the perfect business plan.
The companies that figure out rapid experimentation while keeping quality high and users happy will be the ones that define what tech looks like next.
Hit reply and let me know what you're seeing in your world.
Till next time,
Dev Chandra
CEO @ Startup Intros
Associate @ Context VC
LinkedIn: /in/devchandra
Tim Hsia
Investor @ Context VC
Co-Founder @ Startup Intros
LinkedIn: /in/timhsia
P.S. Raised this week, and we missed you? Want to be featured? Have tips or funding questions? Reply or DM us as we’re here to help.
Startup Intros Weekly Download: Your trusted source for founder-investor insights, delivered with clarity and focus.
