@GoogleDeepMind This is wild. Makes you realize the next real wave of AI won’t just live on screens—it’ll move through the physical world. Robotics is where AI gets its body, not just its brain. Real-time feedback, real-world stakes. That’s when things get interesting.
Kicking off Cohort 9 at Beta University tomorrow — can’t wait.
Every cohort reminds me why we started this: to back scrappy, mission-driven founders before anyone else does.
This one’s special. New energy, new ideas, same hunger.
Let’s build. 🚀
#startups#founders#earlystage
Better to hire and fund people who are scrappy and hungry, who can build, who have high velocity, than the person with the fancy resume with the huge ego
YC has *never* had success funding or hiring the latter
"Hire for slope, not y-intercept." —@paultoo
@karpathy I’m guessing it’s because it want to seem random without actually being random.
I remember reading that odd numbers get picked way more often. Maybe 27 feels just random enough. And AI is mimicking humans. But why do all other models follow this is interesting.Used same training?
7/
The best founders don’t “guess” product-market fit.
They uncover it.
At @UofBeta , we guide builders through market validation, MVPs, and fundraising—fast.
Let the market pull you forward 🧭
#buildinpublic#betau
🧵 Here’s how to find startup ideas the market actually wants.
Most founders build too fast and validate too late.
Let’s fix that ↓
#startups#productmarketfit#buildinpublic
6/
Let data speak.
Track:
CAC (are you burning $$$?)
Churn (do they come back?)
NPS (do they love it?)
Great founders listen twice: first to customers, then to data.
#founderframework#startupmetrics
Totally agree — though I’ve also found that sometimes the obvious path still hides edge if you look closely enough. It’s less about being different for the sake of it, and more about finding quiet advantages where others aren’t paying attention.
@AndrewYNg Nailed it. The best GenAI engineers today don’t just code faster — they compose systems with LLMs, RAG, agents, evals, etc. What separates them is not just speed, but product instinct and the habit of always staying sharp. We’re seeing this firsthand at Beta.
Been lucky to build alongside some insanely sharp founders these past 8 weeks.
From half-broken demos to real momentum — it’s all happening.
If you're in the Bay, come see what we’ve been building.
10 founders. 10 bold ideas.
Live from Silicon Valley.
Join us for Beta Q2 Demo Day — where pre-accelerator startups go from idea → pitch.
📅 June 19, 6PM PT
📍 Palo Alto
🔗 RSVP → https://t.co/JcQcEvJc50
No fluff. Just real momentum.
#DemoDay#Startups#PitchNight#TechEvents
One of the best startup heuristics I've seen:
If it seems like a bad idea but is actually a good one, you're likely onto something big.
Most miss this because they stop at perception.
Takes courage to explore what doesn’t make sense to everyone else-but that’s usually the point
The best investors are high agency with high conviction. Both are different measures of independence.
That’s the only way you can hit the sweet spot of outsize contrarian venture returns.
🚨 Startup Pitch Night is THIS WEEK!
Big ideas, real feedback, and the chance to connect with founders & VCs — all in one evening.
Hosted by @UofBeta & @mirworkai
🗓️ June 5th | 📍 Palo Alto | ⏰ 6:00 PM
🧵 Here’s what’s going down: