... chip supply chain runs through them, regardless of who wins the model race), and Nvidia are the more "obvious but still underpriced relative to trajectory" bets most people point to. The harder question is picking the picks-and-shovels plays before everyone else does infrastructure/chip/cloud companies tend to be less hyped than the flashy AI-native names but often have steadier long-term upside. Comment from Gal Dayan(@galdayan): @burak_emre_taser  that last question ...

... Gowtham Shankar(@gowthamshankar): I m a developer, so I try to keep my subscriptions focused only tools that help me ship faster or stay mentally organized. My current paid stack: ChatGPT (daily dev + writing help) Cursor AI (love the VS Code-like AI experience) ClickUp (lightweight project tracking for solo/small teams) Framer (super quick for landing pages) Notion AI (for async docs with my remote team) Before subscribing, I always ask: Will this save me time or cognitive load? Will ... ... Howell(@howell4change): A few but not a complete list and @OpenAI @Figma @Gamma (super useful) @Fitbod Weight Lifting Workout (a truly great product) @usebubbles , can't live without, I like the permissions and the way the notes are presented vs

... example (UI/UX wise) But struggle to get people paying anything. On top of that if you share your hard work online most people are very hostile towards you (ex. reddit, twitter) because the media has portrait tech founders like shovelling money in their bank vaults. We should probably have a founder only social network where VC funded startups are not allowed :D Comment from Tasos V(@cryptosymposium): @andres_vlaeminck  makes sense. I had 1% with my latest product ...

... they can either be acquired, and if they have the appetite for it, usurp a current contender in that niche space. Time will tell. Comment from Stephen Chow(@growth_journal): It seems the real winners are the "picks and shovels" players. Everyone depends on TSMC to make chips, whether that's Nvidia GPUs or Magic 7's ASICs, and TSMC s key suppliers like Lam Research and ASML benefit as well. Shortly, the cost of AI inference will drop rapidly ...

... mentioned him. I find it interesting that a big part of Marc s success comes from ShipFast and CodeFast, essentially a course on how to build and launch profitable apps. It reminds me a bit of the selling the shovels approach, where instead of just building, you also teach others how to do it. That being said, I totally respect it. He s built a strong brand, and I still find his journey really inspiring. I find it fascinating ...

... Cursor becoming the quickest app in history to $100M . But broadly, I think that means the winners here will be Open AI/Meta/Amazon/Nvidia/Google/Microsoft as they're the funnel for ad-spend/compute and are broadly turning into the picks and shovels for the next generation of software. Some other stuff I think about: How do ads get integrated into LLMs? Is this another $T of market cap for someone or a non-starter? How does everyone becoming 10x more productive impact ...