Skip to main content

​
Environment Variables

The platform requires specific environment variables to function properly. Create a .env file in the root be directory.
1

Create .env file

In the be directory, create a new file named .env:
touch .env
2

Add required variables

Add the following environment variables to your .env file:
.env
MISTRAL_API_KEY=your_mistral_api_key_here
SECRET_KEY=your-secret-key-here
JWT_SECRET_KEY=your-jwt-secret-key-here
MISTRAL_MODEL=mistral-large-latest

​
Environment Variables Reference

​
MISTRAL_API_KEY
string
required
Your Mistral API key for AI-powered features. Get your API key from the Mistral Console.
​
SECRET_KEY
string
required
Flask application secret key for session management. Generate a secure random string.Default: your-secret-key-here (change this in production)
​
JWT_SECRET_KEY
string
required
Secret key for JSON Web Token authentication. Should be different from SECRET_KEY.Default: your-jwt-secret-key-here (change this in production)
​
MISTRAL_MODEL
string
The Mistral model to use for AI responses.Default: mistral-large-latest
Never commit your .env file to version control. Ensure it’s listed in your .gitignore file.

​
Database Configuration

The platform uses SQLite as its database, configured automatically through Flask-SQLAlchemy.

​
Database Settings

The following database settings are defined in backend/config.py: 
SQLALCHEMY_DATABASE_URI = 'sqlite:///interview_prep.db'
SQLALCHEMY_TRACK_MODIFICATIONS = False

​
Database Initialization

The database is automatically initialized when you first run the backend application. The database file will be created at instance/interview_prep.db.
No manual database setup is required. When you run the backend server for the first time:
  1. The backend/instance directory is created automatically
  2. The SQLite database file interview_prep.db is initialized
  3. All required tables are created based on your models

​
First-Time User Setup

The database starts fresh with no users. You must Sign Up a new user account through the application UI to begin using the platform.

​
Configuration Files

The platform uses several configuration files stored in the config/ directory:

​
Topic Rules Configuration

Location: config/topic_rules.json Defines rules and mappings for categorizing interview questions by topic.

​
Taxonomy Configuration

Location: config/taxonomy.json Defines the hierarchical structure of topics and subtopics for the knowledge base.

​
Additional Settings

​
File Upload Configuration

The following settings control file uploads (resumes, documents):
​
UPLOAD_FOLDER
string
Directory for storing uploaded files.Default: uploads
​
MAX_CONTENT_LENGTH
integer
Maximum file upload size in bytes.Default: 16777216 (16 MB)

​
JWT Token Settings

​
JWT_ACCESS_TOKEN_EXPIRE
timedelta
Expiration time for JWT access tokens.Default: 24 hours

​
RAG System Paths

The Retrieval-Augmented Generation (RAG) system uses the following path configurations: 
RAG_DIRS = [
    PROJECT_ROOT / "data" / "processed" / "faiss_mistral",
    Path("data") / "processed" / "faiss_mistral"
]

INDEX_CANDIDATES = [
    "faiss_index_mistral.idx",
    "index.faiss",
    "faiss_index_mistral.faiss",
    "faiss_index.idx"
]

METAS_CANDIDATES = [
    "metas.json",
    "metas.jsonl",
    "metas_full.json"
]
These paths are used to locate the FAISS vector index and metadata files. See Data Indexing for more details.
Configuration complete! Next, proceed to Data Indexing to build your knowledge base.

Build docs developers (and LLMs) love

Get started for freeTalk to us