Vector AI

The AI Business Command Center

Vector is a multi-agent SaaS platform designed to orchestrate a team of executive AI agents. Whether you need startup ideation, technical architecture planning, financial modeling, or go-to-market strategies, Vector puts a full AI C-suite at your fingertips.

Built with a lightning-fast Next.js 16 frontend and a robust FastAPI backend powered by the Google Agent Development Kit (ADK) and Gemini 2.5 Flash.

• Multi-Agent Executive Council: Collaborate with specialized AI roles:
  • 🧠 Prism (Lead Architect): Master orchestrator that delegates tasks and merges outputs.
  • 💼 Atlas (CEO): Business strategy and growth planner.
  • 💻 Nexus (CTO): Technical architecture and feasibility analysis.
  • 📢 Vanguard (Marketing): Brand positioning and launch strategy.
  • 📈 Ledger (Finance): Financial intelligence and monetization strategy.
• Lightning Fast UI: Glassmorphism dashboard with real-time UI updates, dynamic agent status cards, and buttery-smooth Framer Motion animations.
• Automated Document Engine: Instantly generates production-ready markdown documents (Executive Summaries, Architecture Specs, GTM Plans) seamlessly bypassed for 100% reliable local demos.
• Real-Time Telemetry: Live terminal logging system tracking the granular thought processes of the AI agents.
• Persistent Memory: Connects to MongoDB Atlas for persistent project records across sessions.

Frontend (/vector-app)

• Framework: Next.js 16 (App Router, Turbopack)
• Styling: Tailwind CSS, Framer Motion, Glassmorphism UI
• Components: Lucide React, React Markdown, remark-gfm
• Deployment: Vercel Ready

Backend (/mcp-server)

• Framework: FastAPI (Python 3.10+)
• AI Engine: Google Agent Development Kit (ADK), Gemini 2.5 Flash
• Database: MongoDB Atlas (motor asyncio)

• Node.js 18+
• Python 3.10+
• MongoDB Atlas Cluster
• Google Gemini API Key

Navigate to the server directory and configure environment variables:

cd mcp-server
cp .env.example .env

Add your credentials to .env:

MONGODB_URI="mongodb+srv://<USERNAME>:<PASSWORD>@<CLUSTER>.mongodb.net/vector_db?retryWrites=true&w=majority"
GEMINI_API_KEY="<YOUR_GEMINI_API_KEY>"

Install dependencies and start the API:

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python -m uvicorn app.fast_api_app:app --host 0.0.0.0 --port 8000 --reload

Open a new terminal and prepare the Next.js app:

cd vector-app

Create a .env.local file:

MONGODB_URI="mongodb+srv://<USERNAME>:<PASSWORD>@<CLUSTER>.mongodb.net/vector_db?retryWrites=true&w=majority"
MCP_SERVER_URL="http://127.0.0.1:8000"

Install packages and run the development server:

npm install
npm run dev

Visit http://localhost:3000 and enter the Board Room!

The frontend is fully optimized for Vercel deployment.

1. Push your code to a GitHub repository.
2. Go to your Vercel Dashboard and click Add New... > Project.
3. Import your GitHub repository.
4. Set the Framework Preset to Next.js.
5. Set the Root Directory to vector-app.
6. Add the Environment Variables (MONGODB_URI and MCP_SERVER_URL).
7. Click Deploy!

(Note: The backend FastAPI service should be hosted separately on Render, Heroku, or Google Cloud Run, and its public URL provided as the MCP_SERVER_URL).

This project is licensed under the MIT License - see the LICENSE file for details.