Overview

Market Compass is an AI-powered multi-market stock analysis platform currently under active development. It enables traders and investors to analyze stocks across NSE (India) and NASDAQ (USA) using real-time market data, automated technical scanners, interactive charts, and intelligent trading insights.

Instead of manually scanning hundreds of charts every day, Market Compass continuously processes market data, detects high-probability technical setups, and presents actionable opportunities through an intuitive dashboard.

The platform follows a modern API-first architecture consisting of a React/Next.js frontend, a FastAPI backend, PostgreSQL for persistent storage, APScheduler-powered background jobs, and custom Python scanning engines. Every market operates through its own isolated pipeline, allowing the application to support multiple exchanges without code duplication.

Key Features

Multi-Market Support

• Seamlessly switch between NSE (India) and NASDAQ (USA).
• Every dashboard, chart, scanner, and API automatically adapts to the selected market.
• Complete market isolation including symbols, trading calendars, currencies, and data pipelines.

Automated Technical Scanners

Currently includes 15+ automated scanners, including:

Reversal Patterns

• Double Bottom
• Double Top
• Triple Bottom
• Triple Top
• Ascending Triangle

Breakout Detection

• Weekly Breakouts
• Monthly Breakouts
• Yearly Breakouts
• Breakdown Detection

Momentum Scanners

• Top Gainers
• Top Losers
• Gap Up
• Gap Down
• Volume Breakouts

Live Intraday Scanners

• SMA 44 Bounce (5 Minute)
• SMA 44 Bounce (15 Minute)
• Live Moving Average Strategy Detection

Interactive Charts

• High-performance candlestick charts using AG Charts.
• Multiple timeframe support (1D, 1W, 1M, 1Y).
• SMA overlays and technical indicator visualization.
• Planned TradingView advanced chart integration.
• Responsive desktop and mobile experience.

Authentication & User Management

• JWT Authentication
• Google OAuth
• Secure user sessions
• Premium feature access (planned)

Automated Alert System

• Daily scanner reports.
• Live intraday trading alerts.
• Email notifications with embedded stock charts.
• Personalized watchlist alerts.

Tech Stack

Frontend

• Next.js
• React
• TypeScript
• Tailwind CSS
• shadcn/ui
• Radix UI
• Framer Motion
• TanStack React Query
• AG Charts
• React Router

Backend

• FastAPI
• Python
• SQLAlchemy
• Pydantic
• APScheduler

Database

• PostgreSQL
• Prisma ORM
• Redis (Caching)

Market Data

• Yahoo Finance API
• NSE Market Data
• Custom Technical Analysis Engine

Deployment

• Vercel
• Render
• Docker

Architecture

Market Compass follows a decoupled, API-driven architecture.

                React / Next.js Frontend
                          │
                          ▼
                FastAPI Backend Gateway
                          │
          ┌───────────────┼────────────────┐
          │               │                │
          ▼               ▼                ▼
   PostgreSQL        Redis Cache     Authentication
          │
          ▼
    APScheduler Jobs
          │
          ├── Market Data Ingestion
          ├── Technical Scanner Engine
          ├── Pattern Detection
          └── Alert Generation

Multi-Market Pipeline

Each market maintains its own:

• Symbol universe
• Exchange configuration
• Trading calendar
• Timezone
• Currency formatting
• Scanner execution
• Cached datasets

This architecture allows new markets to be added with minimal changes to the application.

Major Challenges

Building a Multi-Market Architecture

Supporting multiple exchanges meant handling different trading sessions, currencies, symbol formats, and market holidays while keeping the application scalable.

Solution

Designed a market-aware architecture where every API request, scanner, and database query receives a market context. This keeps each market completely isolated while sharing the same application infrastructure.

Designing Accurate Technical Scanners

Financial pattern recognition requires balancing sensitivity and accuracy to minimize false positives.

Solution

Developed custom mathematical detection algorithms using pandas, NumPy, and vectorized computations. Pattern detection includes configurable tolerances, confirmation rules, and confidence scoring for improved reliability.

Real-Time Market Processing

Processing thousands of stocks across multiple scanners while maintaining responsive APIs required careful optimization.

Solution

Implemented scheduled background processing, database indexing, intelligent caching, and optimized scanning pipelines to keep API responses fast while continuously updating market data.

Building a Responsive Trading Dashboard

Displaying dense financial information on both desktop and mobile devices presented significant UI challenges.

Solution

Built a responsive dashboard with adaptive layouts, lazy-loaded charts, reusable components, skeleton loaders, and optimized rendering to ensure a smooth user experience.

Performance Optimizations

• React Query caching to reduce unnecessary API requests.
• Background market scanning using APScheduler.
• Optimized SQL queries with database indexing.
• Lazy-loaded routes and components.
• Memoized React components.
• Efficient market-specific caching using Redis.
• Vectorized numerical computations with pandas and NumPy.
• Sequential market ingestion to reduce API rate limits.

Security

• JWT Authentication
• Google OAuth
• bcrypt password hashing
• Input validation with Pydantic
• SQL injection prevention through SQLAlchemy ORM
• Environment-based secret management
• CORS protection
• Premium feature authorization middleware

Future Improvements

• AI-powered trade recommendations.
• Machine Learning pattern prediction.
• Head & Shoulders detection.
• Wedge and Flag pattern scanners.
• Portfolio management.
• Strategy backtesting engine.
• Paper trading.
• Broker API integration.
• Push notifications.
• Mobile applications.
• Additional markets including NYSE, LSE, SGX, and TSE.
• CI/CD deployment pipeline.
• Comprehensive automated testing.

What I Learned

• Building production-ready financial applications from scratch.
• Designing scalable FastAPI architectures.
• Creating efficient stock scanning algorithms.
• Working with real-time market data pipelines.
• Implementing advanced technical analysis using pandas and NumPy.
• Optimizing React applications handling large datasets.
• Managing global application state across multiple markets.
• Integrating scheduled background jobs with APScheduler.
• Building responsive analytics dashboards with modern frontend technologies.
• Deploying full-stack applications using Vercel, Render, Docker, and PostgreSQL.

Project Status: 🚧 Under Active Development

Market Compass is currently a solo project under continuous development. New scanners, AI-powered analysis, additional markets, portfolio management, and advanced trading features are being added incrementally as the platform evolves.