Projects

Core Features

• AI Taste Test: 7-question survey analyzes user preferences to recommend personalized festivals
• Google Calendar Integration: Manage festival schedules with full OAuth 2.0 integration
• Interactive Maps: Google Maps API integration for location visualization and nearby accommodation search
• Real-time Weather: OpenWeather API integration showing weather for 8 major cities
• Social Authentication: Google, Kakao OAuth login support
• Review System: Write and manage festival reviews with media uploads

Technical Stack

Frontend Framework & Build Tools:

• React 19.2.0: Latest version with functional components and Hooks (useState, useEffect, useMemo, useCallback)
• Vite 7.2.4: Ultra-fast build tool with ES Modules and Hot Module Replacement
• React Router DOM 7.12.0: Client-side routing with BrowserRouter

Styling & UI:

• Tailwind CSS 4.1.18: Utility-first CSS framework with custom theme extensions
• Shadcn/ui Components: Radix UI-based accessible component library
• Lucide React 0.562.0: SVG-based icon library

State Management & Data:

• Zustand 5.0.10: Lightweight state management with localStorage persistence
• date-fns 4.1.0: Modern date utility library

Key Libraries:

• FullCalendar 6.1.20: Interactive calendar with drag & drop support
• @googlemaps/js-api-loader 2.0.2: Google Maps JavaScript API integration
• @react-oauth/google 0.13.4: Google OAuth 2.0 authentication

Key Pages & Features

• Landing Page: Immersive video introduction with fade animations
• Taste Test (7 Questions): Companion, atmosphere, important elements, food preference, color, time, participation level
• Festival List: Advanced filtering (region, duration, price, weekend) with real-time search
• Calendar: Google Calendar sync with event CRUD operations
• Map: Festival locations with nearby accommodation search using Places API
• Plan & Curation: AI recommendations with multi-day trip planning
• My Page: User profile, taste analysis, liked festivals, and reviews

Architecture Highlights

• Component-Based Design: Reusable UI components (TownCard, Header, WeatherWidget)
• Performance Optimization: useMemo/useCallback for expensive calculations, code splitting with Vite
• API Integration: OpenWeather API, Google Maps/Calendar/Places API, Kakao OAuth
• Responsive Design: Mobile-first approach with Tailwind CSS breakpoints
• Type Safety: JSDoc comments for type hints without TypeScript overhead

Development Highlights

• Modern Build Tooling: Vite 7.x provides instant HMR and optimized production builds with code splitting and tree shaking
• Component Architecture: Custom reusable components (TownCard, WeatherWidget, Header variants) following single responsibility principle
• Advanced State Management: Zustand with localStorage persistence for user preferences, liked festivals, and taste test results
• API Integration Patterns: Comprehensive OAuth 2.0 flows (Google, Kakao), RESTful API consumption, and real-time data fetching
• Performance Optimization: React.memo, useMemo, useCallback for render optimization; lazy loading for route-based code splitting
• CSS Architecture: Tailwind CSS with custom theme configuration, responsive breakpoints (sm/md/lg/xl), and utility-first approach
• User Experience: Accessibility considerations, loading states, error boundaries, and smooth animations using CSS transitions
• Development Workflow: ESLint configuration, component-based file structure, clear separation of concerns (components/pages/data/lib)

Technical Implementation Details

Google Calendar Integration:

• OAuth 2.0 authentication flow with @react-oauth/google library
• Full CRUD operations on calendar events using Google Calendar API v3
• Token management and refresh handling

Map & Location Features:

• Google Maps JavaScript API with @googlemaps/js-api-loader
• Custom markers for festival locations with info windows
• Places API integration for nearby accommodation search
• Geolocation-based user positioning

AI-Powered Recommendations:

• 7-question taste test analyzing user preferences across multiple dimensions
• Preference-based festival matching algorithm
• Personalized curation with multi-day trip planning capabilities

Data Management:

• JSON-based festival data with comprehensive metadata (dates, locations, categories, prices)
• Client-side filtering and search with real-time updates
• LocalStorage for persistent user data across sessions

Module 1: "Chronos" Datalake Governing the time of data

Goal: Build an automated, high-quality data pipeline that becomes the foundation for all quantitative research (emphasizing data-engineering skill).

Key capabilities:

• Automated ingestion. Use Airflow to automatically collect historical market and on-chain data every day and store it in object storage such as S3.
• Data quality management. Integrate Great Expectations into the pipeline to automatically validate missing values, outliers, and schema, and generate human-readable reports.
• Efficient storage & access. Leverage Parquet and DuckDB so that even very large datasets can be loaded in seconds for any requested time window and shape.

Module 2: "Hermes" Real-time Signal Engine Delivering the market's signals

Goal: Receive market data in real time, analyze it, and generate meaningful trading signals with minimal latency (emphasizing ML engineering / algorithmic trading skill).

Key capabilities:

• Asynchronous data processing. Use asyncio and WebSocket to ingest data from multiple exchanges simultaneously in a non-blocking fashion.
• Real-time feature extraction. Use Redis as a state store and compute rolling microstructure features (e.g., OFI, realized volatility) directly on the stream.
• Message-queue architecture. Employ Redis Pub/Sub to decouple ingestion feature extraction signal generation, ensuring scalability and robustness.

Module 3: "Athena" Strategy & Insight Dashboard Finding wisdom in data

Goal: Discover new alpha strategies using the datalake and signal engine, and analyze/monitor their performance in an intuitive way (emphasizing quant strategy skill).

Key capabilities:

• Regime-based strategies. Use ML models such as HMM to classify the market into regimes (trend / range / high-vol) and automatically switch strategies per regime.
• Systematic experiment management. Adopt MLflow to record parameters, performance metrics, and result charts for a large number of backtests, enabling rigorous comparison.
• Interactive dashboard. Build Streamlit/Dash dashboards to monitor current regime state, recommended strategies, paper-trading performance, and key on-chain indicators.