Member-only story
Vibe Coding is How We Code Now.
“There’s a new kind of coding I call ‘vibe coding’, where you fully give in to the vibes.” — Andrej Karpathy
In early 2025, AI researcher and Tesla’s former director of AI, Andrej Karpathy, coined the term “vibe coding” to describe a radical shift in how developers interact with code. Instead of meticulously crafting code line by line, vibe coding encourages developers to offload most of the creation process to large language models (LLMs), focusing more on what they want to build rather than how.
What Is Vibe Coding?
Vibe coding is the act of using an LLM (like GitHub Copilot, Claude, Cursor, or ChatGPT) to generate entire codebases from high-level prompts without necessarily reading or reviewing the entire output. It’s fast, intuitive, and often driven by experimentation rather than rigorous planning.
Key Characteristics:
- Prompt-based: You describe the functionality, the model generates code.
- Iterative: Errors are fed back to the AI, and the model adjusts.
- Trust-first: Developers often copy-paste code with minimal edits.
- Exploratory: Useful for side projects, prototypes, or creative hacking.
