vibe coding is real. so is the mess it leaves behind. - By Sourav Mishra (@souravvmishra)
okeyy, so let's see what vibe coding really is, why everyone is doing it, and why you still need human eyes on your code.
vibe coding stopped being a joke. you just describe what you want in natural language, and the AI model writes the code.
karpathy nailed it: give in to the vibes and forget the code even exists.
in this post, i, sourav mishra, break down what vibe coding actually is, where it works super well, and why someone still has to own architecture and security.
what vibe coding actually is
vibe coding is building apps by talking to an AI model instead of typing out every line of code by hand.
AI does the typing; you do the directing.
you prompt tools like claude code or cursor. the model drafts the features, and you steer with edits and reviews.
the upside is huge: you ship features super fast and skip boring boilerplate. the catch: if nobody reads what the AI wrote, nobody understands how the app works when something breaks.
where it works—and where it breaks
vibe coding works amazingly well when:
- you're starting a new feature from scratch.
- you're building simple CRUD APIs.
- you're integrating well-documented libraries.
it gets messy when:
- your codebase is huge and full of legacy code.
- the model makes bad assumptions about security.
- nobody reviews the diffs before pushing to main.
what goes wrong
if you just trust the vibes without checking the code, three bad things happen:
- zero ownership: when a bug happens in production, nobody knows where the code lives.
- security holes: AI loves leaving hardcoded secrets or unsafe endpoints.
- code clutter: messy, repetitive code that is hard to maintain.
vibe coding is an awesome superpower, but the model doesn't own production—you do. write prompts, let the AI draft, but always review before you ship!
- Vibe coding = natural-language-driven dev; AI generates, you direct and review. It's real—YC and surveys show teams shipping with 95%+ AI-generated code in segments.
- Upside: faster iteration, less boilerplate. Risk: loss of ownership and security if nobody reads what the AI wrote.
- Where it works: well-scoped tasks, good context, greenfield or clear refactors. Where it doesn't: underspecified tasks, huge messy codebases, architecture-from-scratch.
- Ownership: someone must own architecture, security, and review. Human review, tests, and clear responsibility—you ship; the model doesn't. Tools: Claude Code vs Cursor; agentic chatbot for building your own.
Written by Sourav Mishra. Full Stack Engineer, Next.js and AI.
Frequently Asked Questions
Q: What is vibe coding? Developing by describing intent in natural language and letting AI generate or edit code—minimal manual typing. "AI does the typing; you do the directing."
Q: Is it just a meme? No. YC and survey data show teams shipping with mostly AI-generated code. It's a real segment, especially in early-stage and greenfield work.
Q: What about code quality? AI introduces bugs and security issues. Keep human review, tests, and clear ownership of architecture and security. You ship; the model doesn't.
Q: Which tools should I use? Claude Code for heavier runs and cost control; Cursor for day-to-day editor flow. I compared them in Claude Code vs Cursor. For custom agents, see building an agentic chatbot.