YC Insights.

I spent weeks reading everything YC has ever published about what they look for. here are 9 things that would have changed how I thought about the application

@handotdev with vibe coding we can able to see fortune 5000 companies soon...

overheard from someone in sales: "there are only so many fortune 500 companies"

@Bencera SF got a vibe and environment that push you to test your limits...

Building a startup anywhere other than SF is doing it on hard mode.

@foundr They get confused about what solve problem and how to make money from providing the solution...

The gap between a good idea and a good business is where most people quit.

Claude don't want you to touch the grass this weekend...

@hthieblot Someone should solve the "Enviromental Disaster" related problem

I want to fund as much hardtech as possible this year. Who is building the coolest shit? Bonus for images/videos

I want to fund infinite healthcare apps. AI that delivers more care and less paperwork. Companies like @CounselHealth, @EvidenceOpen and @TennrOfficial are leading the way. If you are building this future, I would love to chat.

@paulg Our mind got cooked, if anything real feels AI generated...

Sometimes real life looks surprisingly AI-generated.

@ycombinator @AgentPhoneHQ @themeetmodi @manav2modi Its time to Build Apple for AI agents...

The $2T telecom industry was built for humans. @AgentPhoneHQ is rebuilding it for agents. One API gives every AI agent its own phone number and a trusted identity to reach the real world. Congrats on the launch, @themeetmodi & @manav2modi! ycombinator.com/launches/QNE-a…

The biggest alpha leak of 2026 is that you can tokenmax $10k/mo with OpenClaw/Hermes + GBrain and get the AI that everyone will have in 2028 for $100/mo, but you can get it now, and that is the biggest single unlock you can have vs your competition

@pmitu Now and may be for future, best is Reddit...

What is the best marketing channel now?

@cb_doge He is such Inspiration to many...

Protect Elon Musk at all costs.

@TTrimoreau Giving prompts...

Tell me one thing you can do that CLAUDE cannot do yet

How YC companies get their first 100 customers 1/ Every YC company that succeeded had to get their first 100 customers. There was no playbook. There was no growth team. There was no marketing budget. Here's how the best ones did it and what you can steal. 2/ Customer 1: Always manual. No YC company got customer 1 from a paid ad, a viral moment, or a press release. Customer 1 came from: → A personal conversation → A cold email to someone who lived the problem → A community the founder was already part of Customer 1 is always earned. 3/ Customers 2–10: The reference expansion. After customer 1: ask them for 3 names. "Who else in your network has this problem as badly as you did?" This referral chain before you have a referral program is the most reliable path to customers 2–10. 4/ Customers 11–30: Community-led growth. Most early-stage YC companies found their next cluster of customers in a community: → A Slack group for their industry → A Reddit community their users live in → A Discord server for their target demographic → A LinkedIn group for their professional audience Show up as a builder, not a marketer. Share what you're building. Ask for feedback. Customers follow. 5/ Customers 31–60: The cold outreach phase. By customer 30, you know exactly who your best customers are. Now: build a list of 500 people who look exactly like them. Cold email with: → One line about the problem → One line about what you built → One line: evidence it works ("22 companies use it, here's what they said") → One specific ask: "Can we get 20 minutes?" Response rate: 5–15% if the list is right. 6/ Customers 61–100: The content channel. The founders who get to 100 customers fastest usually have built an audience. Not a huge one. 500–2,000 people who have the problem your company solves. The best B2B founders write about their problem space not their product. The content attracts the people who need the product. 7/ The YC "do things that don't scale" customer tactics: → Personally onboard every customer via a 1-hour call → Give them your personal phone number for support → Send handwritten notes to your first 20 customers → Fly to meet customers in person in the first 3 months These are exhausting. They don't scale. They generate retention, referrals, and product insights that scale everything else. 8/ The enemy of the first 100 customers: launching broadly too early. "We're launching on Product Hunt next week!" Premature broad launches: → Generate a spike of unqualified users → Overwhelm your support capacity → Confuse your product direction with noise Wait until you understand who your best users are. Then find 100 more of them. 9/ What getting to 100 customers teaches you: → Which users love you (retention insights) → Which users churn (product gap insights) → Which channel brings the best users (growth insights) → What they actually value vs. what you thought they valued (positioning insights) 100 customers is not a revenue milestone. It's a learning milestone. 10/ The "super user" strategy: Among your first 100 customers: 5–10 will be super users. They use the product daily. They send feature requests. They refer others without being asked. Spend 80% of your customer time with these 5–10. What do they love? Why? Build more of that. 11/ The 100-customer milestone and YC: If you have 100 paying customers before applying to YC: → Your application writes itself (you have proof) → Your interview is easier (you can answer "who are your users?" from real data) → Your Demo Day is stronger (investors see a business, not a bet) The first 100 customers is the most important goal before your application. More growth tactics in the full YC playbook, Link in Bio

@paulg make something people want, is still feels surreal, thanks for combining four words that boosted the Moral confidence in today founder...

At one point my son and his friend kept looking for shortcuts to getting rich. Over and over I told them the way to do it is just to make something people want. If this is what I tell my own kids about getting rich, why won't politicians believe this is how a lot of people do it?

@warikoo We are about to see "Warikoo School"

I am shutting down my 100 crores courses business. We’ve grown beyond what I could have imagined, since we started in 2020. 5 lakh students. 100 crores in sales. 25 crores in profits. But it makes no sense to continue it. I’ll be sharing more about my decision and what happens next tomorrow, May 16, 8:30 PM. Set a reminder here: youtube.com/live/VXdxOqjEu…

PerfectBit (@PerfectbitAI) creates high-quality training data that's correct by construction. They verify against physics simulators, scientific databases, formal proof systems. LLMs, robotics, AI for Science, and more. Congrats on the launch, @vaphab and @sjy82155065! ycombinator.com/launches/QMv-p…

PLAN0 (@PLAN0AI) turns architectural plans into construction cost estimates and analytics in minutes. They are building the Bloomberg of construction, with $20B of projects already running through the platform. Congrats on the launch, @abaratiiii, Shervin, and Dimitris! ycombinator.com/launches/QLl-p…

@ycombinator @mdrnhq @sebwpoole @AlexTomovski Wow, so many Companies getting funding...

Modern (@mdrnhq) is building the AI-native operating system for IT, with secure agents that automate help desk, access, devices, security, and on/off-boarding end-to-end. Congrats on the launch, @sebwpoole & @AlexTomovski! ycombinator.com/launches/QII-m…