A fully customizable simulation engine powered by OpenAI, LangChain, and Pinecone. The app guides users through step-by-step decision-making journeys in domains like career, relationships, travel, and more.
- 🔁 Dynamic Simulations: Admin can create reusable simulation templates (e.g. career, travel, etc.)
- 🧠 AI-Powered Guidance: Uses GPT-4/GPT-3.5 for contextual responses based on simulation steps
- 🧩 UI-Aware Responses: GPT adapts to active UI components like sliders, dropdowns, text inputs
- 🧠 Long-Term Memory: Pinecone stores user chat history for personalized and contextual simulations
- 📩 Email Notifications: AWS SES integration for onboarding, alerts, and progress updates
- 🆔 Structured User Tracking: Auto-generated user IDs like #WI0000001 for tracking simulation progress
- 🔐 Security: Token-based sessions, rate limiting, and data retention policies
- 📦 FastAPI Backend: REST API structure for front-end integration (React, Vue, etc.)
- 📁 HTML or JSON Response Support: GPT can return UI-ready HTML or structured JSON for frontend rendering
- Backend: FastAPI (Python)
- AI Engine: LangChain + OpenAI (GPT-3.5 / GPT-4o)
- Memory: Pinecone Vector DB
- Database: MongoDB
- Email: Amazon SES
- Frontend: Any (React recommended)
git clone https://github.com/your-repo/simulation-app
cd simulation-apppython3 -m venv .venv
source .venv/bin/activate # Linux/macOS
.venv\Scripts\activate # Windowspip install -r requirements.txtCreate a .env file:
OPENAI_API_KEY=your_openai_key
PINECONE_API_KEY=your_pinecone_key
PINECONE_INDEX=your_index_name
MONGO_URI=your_mongodb_uri
AWS_REGION=your_ses_region
AWS_ACCESS_KEY_ID=your_aws_access_key
AWS_SECRET_ACCESS_KEY=your_aws_secretuvicorn main:app --reloadServer will be available at http://127.0.0.1:8000
├── main.py # FastAPI entrypoint
├── chains/
│ └── simulation_chain.py
├── utils/
│ └── memory.py
│ └── db.py
├── templates/ # Simulation templates
├── .env # Env variables
├── requirements.txt
└── README.md
Use Postman or Swagger UI to test endpoints like:
/simulate/– Run user simulation/admin/template/create/– Create new simulation template/user/history/– Fetch user chat history