Skip to content

Latest commit

 

History

History
159 lines (129 loc) · 4.74 KB

README.md

File metadata and controls

159 lines (129 loc) · 4.74 KB

🛢️ OG Reasoning Agent

🚀 OG Reasoning Agent is an AI-powered reasoning system for the oil and gas industry, integrating LLMs, structured reasoning (Mind Maps), vector search (Pinecone), and real-time engineering calculations to assist professionals in drilling, wellbore stability, and regulatory compliance.

🌍 Project Overview

This repository implements Agentic Reasoning for oil and gas applications by:

  • 🧠 Mind Map Agent → Structuring knowledge using a graph-based reasoning memory.
  • 🔍 Retrieval Agent (RAG) → Querying Pinecone for drilling manuals, reports, and API standards.
  • 📊 Computation Agent → Performing real-time calculations for mud weight, well control, and casing design.
  • 🤖 LLM Agent → Using DeepSeek R1 / OpenAI for expert-level explanations.

📂 Project Structure

og-reasoning-agent/
│── backend/
│   ├── main.py                # FastAPI entry point
│   ├── agents/
│   │   ├── mind_map.py        # Knowledge graph agent
│   │   ├── retrieval.py       # Web & document retrieval agent
│   │   ├── computation.py     # Engineering calculations agent
│   ├── models/
│   │   ├── embedding.py       # Pinecone vector search
│   │   ├── llm.py             # LLM agent with DeepSeek R1
│   ├── utils/
│   │   ├── config.py          # API keys and settings
│   │   ├── logger.py          # Logging setup
│   ├── requirements.txt       # Python dependencies
│── frontend/
│   ├── src/
│   │   ├── components/        # React components
│   │   ├── pages/
│   │   ├── App.js             # React UI
│   ├── package.json           # Frontend dependencies
│── data/
│   ├── oil_gas_docs/          # Sample documents
│   ├── embeddings/            # Vectorized knowledge base
│── README.md                  # Documentation
│── .gitignore                  # Ignore unnecessary files
│── docker-compose.yml          # Deployment setup
│── Dockerfile                  # Containerization

⚡ Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/ttracx/og-reasoning-agent.git
cd og-reasoning-agent

2️⃣ Set Up Backend

Install dependencies:

pip install -r backend/requirements.txt

Run FastAPI server:

uvicorn backend.main:app --reload

3️⃣ Set Up Frontend

Navigate to frontend and install:

cd frontend
npm install
npm start

🛠 Features

✅ AI-Powered Reasoning

  • Uses DeepSeek R1 for stepwise reasoning and logical deduction.
  • Implements Mind Maps to structure reasoning context.

✅ Retrieval-Augmented Generation (RAG)

  • Searches Pinecone Vector DB for oil and gas drilling documents.
  • Integrates API standards, well logs, and real-time knowledge.

✅ Engineering Calculations

  • Computes mud weight, pore pressure, casing design using SymPy.
  • Supports Python-based real-time calculations.

✅ Multi-Agent Collaboration

  • Mind Map Agent → Structures logical reasoning.
  • Retrieval Agent → Queries documents and API standards.
  • Computation Agent → Executes real-time well calculations.

🔧 API Endpoints

🚀 Query AI Model

Endpoint: POST /query/
Request:

{
  "question": "What’s the recommended mud weight for a well at 12,000 ft TVD with a pore pressure of 0.48 psi/ft?"
}

Response:

{
  "answer": "Recommended Mud Weight: 7.69 ppg"
}

🌍 Deployment

1️⃣ Docker Setup

Build and run containers:

docker-compose up --build

2️⃣ Environment Variables (.env)

Create a .env file and add:

PINECONE_API_KEY=your_pinecone_key
OPENAI_API_KEY=your_openai_key

📚 References

🚀 Contributing

  1. Fork the repository.
  2. Create a new branch (feature/my-new-feature).
  3. Commit your changes (git commit -m "Add new feature").
  4. Push to the branch (git push origin feature/my-new-feature).
  5. Open a Pull Request.

📜 License

This project is licensed under the MIT License.

💡 Future Enhancements

✅ Hybrid Retrieval (Vector + Keyword Search)
✅ Multi-Agent Collaboration
🔲 Support for Real-Time Drilling Data Feeds
🔲 AI-Driven Regulatory Compliance Audits