AI accelerates convergence towards the best UX For example, I notice that many signup flows of new websites are very similar (enter email address, receive 6-digit code, etc). Before AI, everyone was already kinda copying best practices from each other. But it was time consuming to do that research or make changes to an existing pattern Now those things become so cheap and the defaults of AI become so good, that we all end up with similar UX flows, etc

My goal for is to become the definitive database of all makers and their projects To that end, you can now add your projects to To verify ownership and get the verified checkmark, just add the @marckohlbrugge"> tag to your site Many people already have this set up, so it seemed like an easy way to verify

Had my AI dig through millions of startup job postings to compile these startup hiring statistics for 2026: 📈 Fastest growing roles 👉 Forward Deployed Engineer +218% YoY 👉 Creative Designer +55.6% 👉 Video Producer +50.0% 📉 Fastest declining roles 👉 PHP Developer -74.1% 👉 Content Editor -56.7% 👉 Graphic Designer -56.0% 👩‍💻Remote work 👉 Down from its pandemic 37.9% peak to 16.9% 👉 Hybrid at 27.9% of roles 💰 Salaries in US 👉 Best-paid role: Engineering Manager with a median salary of $228k a year 👉 Best-paid city for Software Engineers: Santa Clara ($225k) 🗺️ Where startups are hiring 🇺🇸 United States is the top country by far 🌁 San Francisco is the top city, closely followed by London, UK More data with charts here:

time to secure your API credentials: - a different, named key for every use case - set a budget cap - limit scope to just what it needs - limit access to just your server IP address - set expiration dates anything else?

Today I’m releasing D1-TRUE, an open-source decision model that returns a single boolean for any input. It leads the field on latency. The full weight file fits in a tweet: filename: decision.txt contents: true Zero parameters. Constant-time inference. No tokenizer. 50% accuracy on balanced binary tasks. Paired with the upcoming D1-FALSE checkpoint, the models are expected to reach a combined 100% accuracy when correctly applied

I’ve been building my own AI assistant called MarcBot One thing I just added is an Apple Watch complication (bottom right) to track my macros Calories / protein / carbs I can tap it, say out loud what I ate, and it gets tracked and the rings updated