TrackBets

AI-Driven Financial Intelligence Platform

Overview  •   Tech Stack  •   Features  •   Getting Started  •   Architecture  •   Roadmap  •   Team

---

Overview

TrackBets is a full-stack financial analysis prototype that bridges the gap between complex market data and retail investor decision-making. By synthesizing simulated market feeds with AI-driven sentiment analysis, TrackBets delivers a clear, actionable Buy / Sell / Hold roadmap — all within a premium neon-dark trading interface.

---

Tech Stack

Layer	Technology	Details
Frontend		High-fidelity Neon-Dark UI with Framer Motion, GSAP, Three.js, Recharts
Backend		Hosted on Render · async-first architecture · Uvicorn ASGI
AI Engine		Financial reasoning, sentiment synthesis, and verdict generation
Data		Hybrid: live market feeds + simulated data for demo stability
DevOps		Containerized full-stack deployment with multi-stage build

---

Key Features

Dynamic Analysis Process any ticker (e.g. ZOMATO.NS, TSLA) through a multi-step financial reasoning engine that evaluates fundamentals, technicals, and sentiment signals.	Strategy Wizard User-driven workflow to define entry targets, investment timeframes, and risk tolerance — from Quick Scalp to Long-Term Hold.
Sentiment Synthesis Aggregates news headlines and social media trends via DuckDuckGo Search to provide a real-time Market Pulse percentage score.	Actionable Verdicts Delivers a final Buy / Sell / Hold recommendation with a full AI-generated explanation of the reasoning chain.

---

Getting Started

Prerequisites

Tool	Version
Node.js	≥ 20.x
Python	≥ 3.10
npm	≥ 10.x

1 — Clone the repository

git clone https://github.com/Arthrevs/TrackBets.git
cd TrackBets

2 — Backend setup

# Create and activate virtual environment
python -m venv .venv
.venv\Scripts\activate        # Windows
# source .venv/bin/activate   # macOS / Linux

# Install dependencies
pip install -r requirements.txt

# Configure environment
cp .env.example .env
# Add your GEMINI_API_KEY to .env

3 — Frontend setup

cd frontend
npm install

4 — Run locally

# Terminal 1 — Start the backend
uvicorn api.backend.main:app --reload --port 8000

# Terminal 2 — Start the frontend
cd frontend
npm run dev

Open http://localhost:5173 in your browser.

5 — Docker (alternative)

docker build -t trackbets .
docker run -p 10000:10000 --env-file .env trackbets

---

Architecture

┌─────────────────────────────────────────────────────────┐
│                      CLIENT (React + Vite)              │
│  ┌────────────┐  ┌────────────┐  ┌──────────────────┐   │
│  │  Neon-Dark │  │  Strategy  │  │  Recharts /      |   │
│  │  Landing   │  │  Wizard    │  │  Three.js Viz    │   │
│  └─────┬──────┘  └─────┬──────┘  └────────┬─────────┘   │
│        └───────────────┼──────────────────┘             │
│                        │  Axios HTTP                    │
└────────────────────────┼────────────────────────────────┘
                         │
                         ▼
┌────────────────────────┼────────────────────────────────┐
│                   FastAPI  (Uvicorn ASGI)               │
│  ┌─────────────┐  ┌───┴────────┐  ┌─────────────────┐   │
│  │  /analyze   │  │  brain.py  │  │  scrapers.py    │   │
│  │  endpoint   │──│  AI Logic  │──│  News & Data    │   │
│  └─────────────┘  └─────┬──────┘  └─────────────────┘   │
│                         │                               │
│                         ▼                               │
│               ┌─────────────────┐                       │
│               │  Google Gemini  │                       │
│               │  LLM API        │                       │
│               └─────────────────┘                       │
└─────────────────────────────────────────────────────────┘

---

Project Structure

TrackBets/
├── api/
│   └── backend/
│       ├── main.py            # FastAPI application & routes
│       ├── brain.py           # AI reasoning engine
│       ├── gemini_client.py   # Google Gemini LLM integration
│       └── scrapers.py        # News & market data scrapers
├── frontend/
│   ├── src/
│   │   ├── components/        # React UI components
│   │   ├── hooks/             # Custom React hooks
│   │   ├── services/          # API client layer
│   │   ├── App.jsx            # Root application
│   │   └── main.jsx           # Entry point
│   ├── index.html
│   ├── vite.config.js
│   └── package.json
├── Dockerfile                 # Multi-stage container build
├── requirements.txt           # Python dependencies
└── .env                       # Environment variables (git-ignored)

---

Roadmap

Phase	Milestone	Status
v1.0	Core AI verdict engine + Neon-Dark UI	✅ Done
v1.1	Live API integration (Polygon.io / Zerodha Kite)	🔜 Next
v1.2	Portfolio tracking — real-time P&L via brokerage APIs	📋 Planned
v2.0	Custom ML models for technical indicator prediction	📋 Planned
v2.1	Mobile-responsive PWA with push notifications	📋 Planned

---

Team

Arindam Nath Backend Architect & AI Integration Engineered the FastAPI infrastructure and integrated the Gemini-based LLM reasoning pipeline.

Debanga Shivam Lead Frontend Engineer & UI/UX Designed the high-fidelity React interface, Neon-Dark theme system, and Strategy Wizard workflow.

Meghank Sanjib Dutta Data Scientist & Financial Analyst Specialized in sentiment analysis parameters and financial logic validation of the verdict engine.

---

Built with precision during hackathon crunch time.

_{Made with React · FastAPI · Google Gemini · Vite · Three.js}