Coindcx

🪙 CoinDCX Trading Platform

Sure! Here's a Project Explanation section you can directly add to your README.md (or use separately in your proposal/portfolio):

---

🧩 Project Explanation

🎯 Objective

The CoinDCX Trading Platform aims to create an intelligent and user-friendly crypto trading assistant using the CoinDCX exchange API. It enables users to interact with live market data, manage their crypto portfolio, place/cancel orders, and receive strategic insights — all through a seamless Streamlit web interface and a modular backend powered by CrewAI agents (non-LLM).

---

🛠️ Approach

The platform is designed with a microservice-style architecture, consisting of separate components for API interaction, user portfolio management, order handling, and agent-based market analysis. The app is responsive, modular, and deployable both locally and on Streamlit Cloud.

---

🌟 Key Features Implemented

Feature	Description
Authentication	Simple user login/registration via unique user_id, stored in Redis.
Live Market Viewer	Fetches and filters INR-paired crypto data with last price, volume, 24h change.
Portfolio Tracker	Displays live holdings, locked funds, and estimated INR value using CoinDCX balances + market prices.
Buy/Sell Orders	Place market or limit buy/sell orders through the secure API interface.
Order Management	View current active orders, cancel specific ones, and check trade history.
Agent-Based Analysis	Two rule-based agents simulate human-like logic to detect market trends and advise trades.
FAISS Search Engine	Users can add and search custom strategy/data text using semantic vector retrieval.
Redis Caching	Speeds up repeat data access (e.g., balances, orders) and simulates session-based memory.

---

🧠 How the Platform Works

1. User logs in or registers with a simple user_id (no passwords required). The ID is tracked in Redis for data caching.
2. On login, users can:
   • View filtered live market data from CoinDCX.
   • See a snapshot of their portfolio using account balance + market price.
   • Place buy/sell orders (limit or market type).
   • Check and manage active orders or historical trades.
3. In the "Agent Analysis" tab:
   • A Market Analyst Agent calculates average vs. current price and labels trends as bullish/bearish.
   • A Trade Advisor Agent provides trade suggestions based on market outlook and portfolio data.
   • No LLM is used — logic is 100% interpretable and rule-based.
4. In the FAISS section, users can:
   • Input and vectorize strategy notes or logs.
   • Perform semantic search for related entries.
   • All embeddings are stored locally using sentence-transformers.

---

Let me know if you'd like this turned into a portfolio PDF, presentation slide, or GitHub README version with visuals!

---

🚀 Features

• 👤 User Authentication & Caching – Login/Register using a simple user_id, stored with Redis.
• 📈 Live Market Data – Search and view current INR-based markets with price/volume metrics.
• 💼 Portfolio Viewer – View current holdings, locked balances, and INR value estimations.
• 💸 Trade Execution – Place market/limit buy/sell orders securely.
• 📃 Order Management – View, cancel, and track current or historical orders.
• 🤖 Agent Analysis – Get trading suggestions from rule-based CrewAI agents (no LLM required).
• 🔍 FAISS Integration – Semantic vector search for embedded historical text data (e.g., logs, strategies).
• 🔐 Secure API Handling – CoinDCX API credentials handled with .env and HMAC SHA256 signing.

``

📂 Project Structure

## 📦 Installation
 
 ### 1. Clone the Repository

  git clone [https://github.com/yourusername/coindcx-trading-platform.git](https://github.com/SINGHBP29/Coindcx/edit/main/)
  cd coindcx-trading-platform

2. Setup Python Environment

pip install -r requirements.txt

3. Add API Credentials

Create a .env file in the root directory:

COINDCX_API_KEY=your_api_key
COINDCX_API_SECRET=your_api_secret

These are used securely for authenticated API calls.

---

▶️ Running the App

Streamlit UI (Recommended)

streamlit run app.py

Terminal CLI (Optional)

python main.py

---

## 🔍 AI Agent Analysis (No LLMs Required)

Simulated agents use historical market data to analyze trends and provide suggestions:

- **Market Analyst Agent**: Identifies bullish/bearish patterns using price averages.
- **Trade Advisor Agent**: Recommends actions based on portfolio and market conditions.

No cloud inference or LLMs are needed — logic is purely rule-based and local.

---

## ⚡ Redis & FAISS Integration

- **Redis**: Caches user data (e.g., portfolio, active orders) for fast reloads.
- **FAISS**: Allows semantic search of trading logs, market summaries, or any vectorized text data.

---

```toml
COINDCX_API_KEY = "your_key"
COINDCX_API_SECRET = "your_secret"
```

5. Deploy and share the public link 🎉

---

✅ TODO / Enhancements

• Add chart-based portfolio visualization
• Webhooks / Telegram alerts for trade events
• User-level analytics with FAISS insights
• Role-based access & multi-user dashboards

---

🧠 Credits

Made with ❤️ using:

• [CoinDCX API]
• [Streamlit]
• [FAISS by Meta]
• [Redis]
• [AI Agents]

---

`

📜 License

MIT License. Use it, fork it, improve it!

---

🙌 Author

**Bhanu Pratap Singh**  
📫 Connect: [https://linkedin.com/in/your-profile](https://www.linkedin.com/in/bpsingh29/) | [https://github.com/yourusername](https://github.com/SINGHBP29)