achawla19/intuitcode-extension — explained in plain English
Analysis updated 2026-05-18
Practice explaining your coding approach out loud before writing any code, like in a real interview.
Get structured Socratic hints on a LeetCode problem without being handed the solution.
Have an AI review your finished code for bugs and complexity once you've written it.
Move through clarify, brute force, optimize, and code review phases that mirror a real technical interview.
| achawla19/intuitcode-extension | 4ssh1/mine-sweeper | adun-denton/chinvat | |
|---|---|---|---|
| Stars | 2 | 2 | 2 |
| Language | TypeScript | TypeScript | TypeScript |
| Last pushed | — | 2026-01-31 | — |
| Maintenance | — | Maintained | — |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 4/5 | 2/5 | 4/5 |
| Audience | developer | vibe coder | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires cloning a separate backend repo, running it locally, and providing a free Groq API key.
IntuitCode is a Chrome browser extension that turns practicing LeetCode coding problems into something closer to a real technical job interview. Instead of jumping straight into writing code, you first explain your approach out loud, and an AI interviewer listens, asks clarifying questions, gives hints without spelling out the answer, and later reviews your finished code, the way an actual interviewer would. The extension adds a panel on top of any LeetCode problem page that automatically picks up the problem's title, difficulty, and tags. You click a button to start speaking, and the AI responds with a question or a hint rather than a solution. The conversation is structured into four stages that mirror a real interview: clarifying the problem, working out a brute force approach, optimizing it, and finally reviewing your actual code once you write it. At the end, you get feedback covering time and space complexity, what you did well, and suggestions for similar problems to practice next. While you can type instead of speaking, the README is clear that voice is the point of the tool, since real interviews require you to reason out loud under pressure rather than type at your own pace. Your spoken words show up as an editable transcript, so you can fix anything the speech recognition misheard before sending it. Hints follow a three level system designed to nudge you toward the answer without ever giving it away directly. Under the hood, the extension is built with TypeScript on Chrome's Extension Manifest V3, uses the browser's built in speech recognition and speech synthesis features for voice, and talks to a separate backend server built with Node.js and Express, which is what actually calls the Groq AI service running a Llama model. The extension itself never holds the AI API key directly, all requests are routed through that backend so the key stays hidden from anyone inspecting the extension's code. Setting it up for local development means cloning both the extension and backend repositories separately, installing dependencies, building the extension, and loading it as an unpacked extension in Chrome, while the backend needs its own free Groq API key. The project is released under the MIT license.
A Chrome extension that turns LeetCode practice into a voice-based mock interview with an AI that hints instead of giving answers.
Mainly TypeScript. The stack also includes TypeScript, Chrome Extension, Node.js.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.