Another common question I’m answering working with scaling tech companies is… Q. What’s the worst leadership advice you’ve heard? A. By far the worst is “Hire great people and get out of their way”. Let me explain… 🧵 (1/32)

Everyone’s busy summoning synthetic minds, and I’m starting my day with a notebook and callipers trying to make atoms behave. AI helped by taking a top-down image of the replacement part and created an SVG to import to Fusion 360 and trace. Onshore manufacturing!

On Grokene, Grok and AIs cannot predict a room temperature ambient pressure superconductor now, because we don't know its superconducting mechanism. It may take 10 - 20 years for Grok and AIs to predict a room temperature superconductor material. Grok said, "The project lives in

I’m crossing the Gulf Stream today and it’d always a wild ride. Weeks of weather analysis to find the perfect window. It transports about 1.3 petawatts of heat northward ~100 times all the energy humans use globally.

👉🏼

Our app is 1 year old 🎂. To celebrate we put together a documentary with inspiring stories from our users that make all the hard work so so worth it. - Rescue at sea leading to a world record - Making lasting memories with severe health prognosis - True friendship

The Q3'25 Power 20 Report is live. We analyzed 50,000+ data points across market signals and commercial traction to rank the most promising pre-IPO names. What we found: The top 5 companies (SpaceX, Anduril, Anthropic, OpenAI, Databricks) averaged 63% share price growth. They

Finally have time to come up for air. Want to share some tips from 2.5 years building on @supabase (supa) in case they are useful for others. 1. The dashboard is cute, but don't use it Most of the docs and demos show devs clicking around in the supa dashboard. It's good for

Early on I thought success = building the perfect product. Then I learned it’s marketing. Then retention. Then monetization. But I'm only good at one of those! The real secret sauce is finding co-founders who are world class at everything you’re bad at. My co-founders are

It's fun to have users test the limits of your product. These adventurous woman rowed across the pacific and tracked and shared their trip entirely on SeaPeople. 147 days and over 8k miles. Mostly offline as they had to conserve battery and starlink is watt hungry and limited

2 years later and from 0 to 100k users and this extremely simple architecture works great. Easy for AI to grok and work in its bounds. Haven’t changed a thing outside of @remix_run renaming. @supabase and @typesense just work and adopting new @reactnative arch this month. @expo

Lately I’ve seen people criticize founders for “lacking nuance.” It’s a fair point, but also a deeply nuanced accusation. You can’t lead by caveat. You can only think with nuance, then decide with clarity. Nuance helps you understand the world. Clarity helps you ship. Most

The Great AI Mispricing Cash is not a moat at the app layer A lot of people have asked if we should have raised $100m+ at Spellbook. We have no need for more cash. We have lots of revenue from 4,000 customers. But mainly, we are wary of a serious market distortion... 🧵

Everyone is dunking on this story. The reality is that monoliths rust and get very complex. You have to invest to scale them. It’s not like the solution was to do nothing. But breaking it apart isn’t the only option. What we’ve done at a few companies to keep a monolith and

Nobody told me that buying a 3D printer is like crossing a one-way bridge. On one side you're a normal person. On the other side you're explaining to your spouse why you need to print a custom soap dispenser at 2am. Spending 12 hours designing a $1 plastic part. Never going back.

Running customer support for your startup teaches you that humans contain multitudes. Same day: 'You changed my life thank you so much!' and 'I hope your company burns to the ground' over the exact same feature. The range of human emotion in my inbox is absolutely unhinged.

The biggest drain on your team isn't the impossible challenges. It's the people who turn simple tasks into complex ones. Keep the simplifiers. Avoid the complicators. It's not easy to spot, often the complicators are seen as having a great attention to detail & high care. But

What a strange goal. It should be 100% ai-written and 100% human rewritten. Not by October, by tomorrow. AI is a genius prototyper, but most of the code should be thrown out. Use as a tracer bullet into your code and idea surface area generator. It will speed you up massively,

@nikitabier We had a spam problem as well in our app. We implemented an auto mute based in our spam signals. Spammers have to blast a large surface area and play the odds game that someone will respond. Thats a huge spam signal. While muted, their chat / replies are removed and the app is

Doubled conversion this month. No new features. Thanks @RevenueCat for paywalls 2.0.

Finally!

I pair-interview with CEOs who are hiring engineering leaders. The top 2 high-selectivity bandpass filters are: - What does a 5x better version of you look like? - Who is 5x better than you that you learn from? - Show me your .cursor/rules files