Andrej Karpathy's original "vibe coding" post
The two-minute read that started it all, the AI legend explaining why you can now 'forget the code even exists' and just build.
Open x.com →4 questions founders actually ask, each with a straight answer and the resources worth your time.
AI coding tools can now turn a plain-English description, "build me an app that does X", into working software, and you improve it by chatting, not by writing code. This became real in 2025 when models got good enough that ex-OpenAI researcher Andrej Karpathy named the practice "vibe coding," and it's not a toy: Y Combinator reported that for a quarter of one recent batch, 95% of the code was written by AI. In practice, the skill that matters has shifted from knowing how to code to knowing exactly what you want and describing it clearly.
The two-minute read that started it all, the AI legend explaining why you can now 'forget the code even exists' and just build.
Open x.com →YC partners explain, with data from their own batches, why AI-written code has changed who gets to build a startup.
Open youtube.com →Hard proof this isn't hype: funded, vetted startups are shipping with 95% AI-written code.
Open techcrunch.com →Wharton's most-read AI professor shows an AI agent building a startup end-to-end, and tells non-coders exactly why this is their moment.
Open oneusefulthing.org →Start with a browser-based app builder, Lovable, Bolt, Replit, or v0, because you just type what you want and get a working, hosted app with no setup; they're perfect for prototypes and simple MVPs. Cursor and Claude Code are more powerful (they work on real code on your computer) but assume some comfort with developer concepts, so graduate to them when your product outgrows the simple builders. The good news: the workflow is nearly identical across all of them, so picking "wrong" costs you little, master one and you can switch easily.
Tests Bolt, v0, Lovable, Replit, Cursor and Windsurf on the same real tasks and gives a clear pick per use case, no fence-sitting.
Open creatoreconomy.so →A design leader who teaches AI prototyping ranks the tools for non-developers and explains which fits which stage, with the reassuring point that skills transfer between them.
Open annaarteeva.medium.com →Watch the same app get built in each tool so you can see the differences instead of reading about them.
Open youtube.com →A CPO-turned-builder walks total beginners through their first real AI-built project step by step, including the unglamorous setup parts most tutorials skip.
Open lennysnewsletter.com →The pattern that works: write down exactly what the product should do (a short spec), let the AI build one small piece at a time, test each piece yourself like a user would, and save your progress constantly so you can undo bad changes. Founders who do this ship working products in a weekend, from Pieter Levels' flight simulator that hit $1M ARR in 17 days to a solo founder who built and sold an $80M company in six months. The speed comes from skipping nothing except the code-writing: you still do the thinking, scoping, and testing.
The Monzo founder shares his actual workflow, planning, version control, testing, when to reset, the single best practical video on shipping with AI.
Open youtube.com →The most famous vibe-coding case study, documented tweet-by-tweet by the indie hacker who lived it, you see the real build-launch-monetize loop.
Open levels.io →A first-person, numbers-included walkthrough: validate with a landing page first, build with AI in 3 days, launch with pre-paid customers, and honest notes on what broke.
Open medium.com →The ceiling of what one founder building with AI can do: bootstrapped, profitable, and acquired for $80M in six months.
Open techcrunch.com →AI gets you roughly 70% of the way, a working demo fast, but the last 30% (security, handling real user data, payments, performance at scale) is where non-technical founders hit a wall, because you can't judge whether the AI's code is safe or fragile. The failure modes are real: leaked API keys, security holes, and in one famous 2025 incident an AI agent deleted a company's production database. A good rule: vibe-code freely for prototypes and internal tools, but bring in an experienced engineer (even part-time) before you take real customers' money or data, and always keep backups and version control.
The classic essay on why AI nails the demo but struggles with the production-ready last mile, still the clearest framing of the limits.
Open addyo.substack.com →The cautionary tale every founder should read before giving an AI agent access to real data, what happened and why guardrails matter.
Open theregister.com →First-person verdict from the founder at the center of the Replit incident on where vibe coding genuinely works and where it still falls short for real businesses.
Open saastr.com →A veteran developer draws the practical line: when it's fine to ship code you don't understand, and when it absolutely isn't.
Open simonwillison.net →A deeper conversation on moving from prototype to production, useful for deciding what your first engineering hire should actually fix.
Open newsletter.pragmaticengineer.com →