Interview-Analyzer is a tiny application designed to help you analyze and improve your job interview performance. It leverages AI to transcribe and analyze your interview recordings, providing actionable feedback.
- AI-Powered Analysis: Automatically evaluates your interview performance using advanced LLMs.
- Transcription: seamlessly converts audio recordings into text for review.
- Historical Analysis: Keeps a record of your past interviews and analyses for tracking progress.
- Secure Access: Simple username/password authentication to protect your data.
- Interactive Feedback: Re-analyze specific parts or ask follow-up questions about your performance.
This project is built using a modern, robust stack:
- Streamlit: For building the interactive web interface.
- LangChain: To orchestrate the LLM interactions and analysis logic.
- PostgreSQL: For persistent storage of transcripts and analysis results.
- Docker: For consistent environment and easy deployment.
- Docker and Docker Compose installed on your machine.
- An OpenAI API Key.
-
Clone the repository:
git clone <repository_url> cd interview-analyzer
-
Configure Environment Variables: Copy the example environment file:
cp .env.dist .env
Open
.envand fill in the required values:OPENAI_API_KEY: Your OpenAI API key.DB_HOST,DB_PORT,DB_NAME,DB_USER,DB_PASS: Database credentials.LOGIN_USER,LOGIN_PASSWORD: Credentials to log in to the app.
-
Start the Application: Run the following command to build and start the services:
docker-compose up --build
- Open your browser and navigate to
http://localhost:8501. - Log in using the credentials you defined in
.env. - Analyze New Interview: Go to the main page to upload an audio file. The system will transcribe and analyze it.
- View History: Use the sidebar to navigate to past interviews and review the AI's feedback.
The project includes Adminer for easy database management.
- Access Adminer at
http://localhost:8080. - System: PostgreSQL.
- Server:
database. - Username/Password/Database: As defined in your
.envfile.