A comprehensive job search management application with:

• Resume management
• Interview question bank
• Job application tracking
• Document upload and semantic search
• Google Doc Integration

Neon.tech (serverless PostgreSQL) + pgvector is used as the single source of truth for all structured data and semantic search.

• docs/ARCHITECTURE.md - Complete architecture overview
• docs/DATABASE_SCHEMA.md - Detailed database schema documentation
• docs/NEON_MIGRATION_PLAN.md - Migration guide for Neon.tech (serverless PostgreSQL)

1. Neon.tech account (free tier available) - Sign up
2. Python 3.10+
3. Google API Key for embeddings

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up Neon.tech database (see docs/PGVECTOR_SETUP.md)
# 1. Sign up at https://neon.tech/
# 2. Create a new project and database
# 3. Get your connection string from the dashboard
# 4. Set NEON_DATABASE_URL environment variable (or use Streamlit secrets):
export NEON_DATABASE_URL="postgresql://username:[email protected]:5432/chat_pgvector?sslmode=require"

# Run migrations (using Neon connection string)
psql "$NEON_DATABASE_URL" -f storage/migrations/001_create_vector_tables.sql
psql "$NEON_DATABASE_URL" -f storage/migrations/002_add_jsonb_indexes.sql

# Start the application
streamlit run app.py

For detailed setup instructions, see docs/PGVECTOR_SETUP.md.

web framework

PostgreSQL + pgvector (vector similarity search)

embedding framework, models: "models/gemini-embedding-001"

Connect LLMs for Retrieval-Augmented Generation (RAG), memory, chaining and agent-based reasoning.

documents import

audio transcription

video processing

Serverless PostgreSQL database for all structured data and vector storage (managed, auto-scaling)