⚡ S2AI — Dual-AI Interaction Playground

S2AI is an open-source experimental project that allows two AI models to interact, reason, and respond together, powered by a React frontend and a Node.js + Express backend.

This project was built to move beyond tutorials and into real system design, provider orchestration, and frontend–backend integration.

• Build something from scratch
• Learn how AI systems are wired together
• Experiment with multiple AI providers
• Understand real-world backend architecture
• Create a usable, open-source learning project

This is not a SaaS, not a product, and not monetized. It’s a learning playground. The Full project is currently under Construction and don't know when realissed

• AI-1 generates a response
• AI-2 reviews / critiques / adjusts internally
• Only the final, public-safe output is returned to the user
• Provider logic is modular and extensible

• React (Vite)
• TypeScript
• React Router
• Tailwind CSS
• Context API (in-memory state)

• Node.js
• Express.js
• Modular services & controllers
• Provider-based architecture

• OpenAI
• Gemini
• Ollama (local models like LLaMA 3)
• Easy to extend with new providers

/frontend
 └─ src
    ├─ app
    ├─ components
    ├─ pages
    ├─ context
    ├─ services
    └─ utils

/backend
 └─ src
    ├─ routes
    ├─ controllers
    ├─ services
    ├─ providers
    └─ app.js

git clone https://github.com/your-username/S2AI.git
cd S2AI

cd backend
npm install
npm run dev

Runs at:

http://localhost:5000

cd frontend
npm install
npm run dev

Runs at:

http://localhost:5173

You must provide your own API key from one of the supported providers:

• OpenAI
• Gemini
• Ollama (local)

Paste the key into the Settings section inside the app.

🔐 API keys are:

• validated by the backend
• kept in memory only
• never stored or logged

• Experimental project
• Prompts and logic are still being refined
• No authentication (by design)
• Not production-ready
• Focused on learning, not scale

• Provider-based system design
• Frontend vs backend responsibility
• API validation strategies
• Real debugging (not tutorial land)
• Why building beats watching

Feel free to:

• Fork the repo
• Improve prompts
• Add providers
• Refactor architecture
• Experiment freely

PRs and ideas are welcome.

Open-source. Use it. Break it. Learn from it.

At some point the question changed from “What tutorial should I watch next?” to “When am I going to build my own thing?”

This is that thing.