BidForge AI 🚀

BidForge is an AI-powered Request for Proposal (RFP) and Bid Automation system. It streamlines the creation of sales bids, proposals, and compliance checks by leveraging LLM capability (OpenAI + Mistral AI fallback) and a Retrieval-Augmented Generation (RAG) knowledge base.

🌟 Key Features

RFP Document Ingestion: Upload RFPs in PDF or DOCX format. The system automatically parses requirements and stores them.
RAG Knowledge Base: Upload company context, past proposals, brochures, and Q&A documents. The text is vectorized and stored in a Chroma DB vector database for semantic search.
AI-Powered Proposal Generation: Generates highly tailored, professional answers to RFP questions based on relevant context retrieved from the Knowledge Base.
Automated Compliance Checking: Evaluates generated proposals against RFP compliance criteria and lists improvements.
Export Capabilities: Export finalized proposals to professional PDF or Word formats.

* User Management: Secure authentication using JWT tokens and encrypted password storage.

📁 Repository Structure

BidForge/
├── backend/          # FastAPI Backend (Python)
│   ├── app/          # API, schemas, models, and core logic
│   ├── uploads/      # Temporary storage for uploaded RFPs
│   ├── exports/      # Staging area for generated PDF/DOCX bids
│   ├── requirements.txt
│   └── .env.example
├── frontend/         # Next.js Frontend (React/TypeScript)
│   ├── src/          # Page routes, components, and API client
│   ├── public/       # Static assets
│   ├── package.json
│   └── tsconfig.json
└── .gitignore        # Global gitignore configuration

🛠️ Tech Stack

Frontend: Next.js (App Router), React, TypeScript, Tailwind CSS
Backend: FastAPI, Uvicorn, SQLAlchemy
Database: PostgreSQL / SQLite
Vector Search (RAG): Chroma DB
LLM Integration: Dual-engine OpenAI GPT Models & Mistral AI Fallback (Mistral Large)
Document Parsers & Generators: PyMuPDF (fitz), python-docx, ReportLab (PDF generator)

* Containerization: Docker (Multi-stage Node.js & Python slim)

🚀 Getting Started

Prerequisites

Ensure you have the following installed on your machine:

Python 3.10+

* Node.js 18+

Backend Setup (FastAPI)

Navigate to the backend directory: bash cd backend
Create a virtual environment: bash python -m venv .venv
Activate the virtual environment:
- Windows: powershell .venv\Scripts\activate
- macOS/Linux: bash source .venv/bin/activate
Install dependencies: bash pip install -r requirements.txt
Configure environment variables: Copy the example environment file and configure your API keys: bash cp .env.example .env Open .env and fill in the values:
- Set OPENAI_API_KEY to your OpenAI Key.
- Update DATABASE_URL to your PostgreSQL database connection string, or use SQLite (e.g., sqlite:///./bidforge.db).
Run the Backend server: bash uvicorn app.main:app --reload The API docs will be available at http://localhost:8000/docs. --- ### Frontend Setup (Next.js)
Navigate to the frontend directory: bash cd ../frontend
Install dependencies: bash npm install
Run the development server: bash npm run dev Open http://localhost:3000 with your browser to see the BidForge application dashboard. --- ## 🐳 Docker & GCP Cloud Run Deployment

Both the backend and frontend services are containerized and can be built/deployed directly to Google Cloud Run:

1. Backend Service

Create and configure environment variables in backend/.env (set MISTRAL_API_KEY for Mistral generation fallback).
Deploy the local folder to GCP: bash gcloud run deploy bidforge-backend --source backend/ --project=YOUR_PROJECT_ID --region=us-central1 --allow-unauthenticated

2. Frontend Service

Create a frontend/.env.production file pointing to your deployed backend URL: env NEXT_PUBLIC_API_URL="https://your-backend-cloud-run-url.a.run.app"
Deploy the frontend folder to GCP: bash gcloud run deploy bidforge-frontend --source frontend/ --project=YOUR_PROJECT_ID --region=us-central1 --allow-unauthenticated