amrutha97

AI mock interviewer for coding interviews. 132+ real-world problems, live code execution, voice mode, and structured feedback.

19
1
100% credibility
Found Mar 29, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

CodePrep is a local web application that simulates realistic coding interviews powered by AI, offering problem selection, code execution, voice support, detailed feedback, and progress tracking to help users prepare for technical interviews.

How It Works

1
🔍 Discover your personal interview coach

You find a helpful tool that runs on your computer to practice coding interviews like the real thing.

2
🚀 Launch it with a click

Open your web browser to see a friendly list of real-world coding challenges.

3
📚 Pick a challenge that excites you

Browse topics, difficulties, or surprise yourself with a random one to start practicing.

4
Choose your practice style
💭
Study mode

Chat with a tutor to understand the problem deeply before trying.

🎤
Interview mode

Talk or type with an AI interviewer who asks follow-ups just like a real job interview.

5
Solve and submit your code

Write your solution in the built-in editor, run tests, and see instant results while the interviewer responds.

6
📊 Receive honest feedback

Get a detailed report with scores, strengths, improvements, and a hire recommendation.

🏆 Track your growth over time

See your progress across problems and categories, ready for real interviews with confidence.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is codingprep?

Codingprep is a local AI mock interview tool for coding prep, simulating real interviews with 132+ real-world problems across categories like stateful systems, parsing, trees, and DP. Pick a problem, chat back-and-forth with a GPT-4o-powered interviewer that probes reasoning and adds constraints, write Python code with live execution and tests, then get structured feedback on code, communication, and hire signals. Voice mode via WebRTC lets you practice talking aloud, all stored locally for privacy—no data leaves your machine except OpenAI calls.

Why is it gaining traction?

Unlike puzzle grinders or static platforms, it mimics live interviews with dynamic follow-ups, mid-session twists, and detailed debriefs (strong hire/no hire ratings plus critiques). The 132+ problems tie to engineering scenarios (LRU caches, rate limiters), filters by difficulty/category, tracks progress with status dots, and supports study mode for pre-interview research. Local runs keep costs low (~$0.10-0.50/session) and sessions resumable.

Who should use this?

Backend engineers grinding LeetCode-style interviews at Big Tech, especially those weak on verbalizing approaches or handling follow-ups. Ideal for mid-level devs prepping system design lite problems (caches, schedulers) or juniors building Python fluency with real test cases. Skip if you need non-Python langs or free tiers.

Verdict

Solid for solo practice—excellent docs, intuitive UI, and thoughtful features like history/resume make it addictive despite 19 stars and 1.0% credibility score. Early but mature enough for daily use; contribute problems via YAML to grow the library.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.