AI Products & Indie Growth · Rising on X

Project Suncatcher has liftoff! ☀️🚀 Announced last year, this moonshot project explores if we can one day host machine learning infrastructure in space. After years of research, we’ve just launched a prototype satellite, built in partnership with @planet, into orbit on @SpaceX's Transporter-18 rideshare mission. This initial mission will gather data on how @Google TPUs handle the physical stress and extremes of spaceflight. Ultimately, these findings will help us explore how to potentially link multiple satellite constellations for scaled machine learning. By operating in low Earth orbit, these systems can tap into near-constant sunlight to generate up to 8x more solar power than on Earth.

Google just launched four of its AI chips into orbit on a Falcon 9, and the whole project hinges on one number buried in its own research paper. At $200 per kilogram to low Earth orbit, launching a data center starts to cost about the same as powering one on Earth. That math sounds crazy until you see what terrestrial data centers actually pay. $570 to $3,000 per kilowatt per year for electricity, plus years stuck in grid interconnection queues before a single server turns on. The satellite that went up today flies a dawn-dusk orbit where its panels almost never see shade, generating up to 8x what the same panel produces on the ground. The sun becomes a power plant nobody has to build. The chip decision is the part worth studying. This satellite carries stock Trillium TPUs, the same commercial silicon running in Google Cloud, instead of radiation-hardened chips that run generations behind the state of the art. Google fired a 67 MeV proton beam at them in the lab, and the memory didn'

The US just launched one AI front door to 29,000+ government websites. So I built one for Pakistan 🇵🇰 Ask in English, Urdu or Roman Urdu. It answers only from official govt sites, with sources. Shared it on LinkedIn, 1,000+ users in few hours 🤯 https://t.co/YJVLm7wUpB https://t.co/kHzgXHXLMp

the internet is officially cooked💀 this team just launched Griffin, the worlds first human interaction model. this AI model creates human like characters, and it’s insanely good 48% of people that interacted with it didn’t even realize that it was AI… 😭 imagine where we will be in the next 6 months.

I met this guy Mike from Australia who runs 5 SaaS apps doing $200K/month combined - fully bootstrapped, no VC, no exits. His whole thesis: “I like to build ideas that can’t fail.” Every app follows the same 10-step playbook. He’s run it on 3 companies, halfway through a 4th (https://t.co/Idk078vzLo), and about to start the 5th (https://t.co/CPqK5GGtRB, a B2B group e-cards tool launching in 2 weeks). The rules he never breaks: > Pick an idea that’s already been done. New ideas need validation. Copied ideas already have proven demand. > Always start with 4 co-founders, split 25% equally. Minimizes founder fallout - the #1 killer of startups. > Tech-heavy team: frontend dev, backend dev, designer, and someone like Mike doing everything else. > Never give an account away free. If they pay, they use it. > Grow to $10K MRR to cover costs, then split all profits between founders. Stay lean forever. The money engine before MRR? Lifetime deals. Sell it for $59-$100 one-time. He raised

$NOW just launched Flow which is an AI native service desk that brings IT agents directly into $CRM Slack and $MSFT Teams with no infrastructure required. You can see where ServiceNow’s moat starts to show up in AI with management saying its “meeting companies where their employees work every day” rather than making them build agents in house.

Someone: don't build in public, people will steal your idea and build it faster. Now Lagos Danfo has launched and suddenly everyone building similar product are out. I have seen nearly 6 more now. One said, "I started building a long time ago before he did." Okay?