Skip to content

Mayank-glitch-cpu/interview-unlocked

 
 

Repository files navigation

Interview Unlocked

An AI-powered interview preparation system using LangGraph and multi-agent orchestration.

Setup

  1. Install dependencies:
pip install -r requirements.txt
  1. Set up environment variables in .env:
GOOGLE_API_KEY=your_google_api_key
OPENAI_API_KEY=your_openai_api_key
TAVILY_API_KEY=your_tavily_api_key
  1. Place your Google Cloud credentials in key.json (needed for speech-to-text)

Usage

  1. Have your resume in PDF format and job description in a text file
  2. Run the script:
python interview_unlocked.py

The system will:

  • Process your resume and job description
  • Generate a personalized study plan
  • Suggest relevant LeetCode problems
  • Generate and ask interview questions
  • Record and evaluate your answers

Files

  • interview_unlocked.py: Main script
  • requirements.txt: Python dependencies
  • .env: Environment variables (create this)
  • key.json: Google Cloud credentials (required for speech-to-text)
  • Leetcode-company-problem-set.xlsx: Company-specific LeetCode problems

About

An AI interview coach built with LangGraph. It processes resumes and JDs, infers company expectations from websearches to generate personalized interview questions. User audio answers are transcribed and evaluated against inferred rubrics, providing detailed feedback to improve interview performance.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Jupyter Notebook 93.1%
  • Python 6.9%