Evan Twidwell - Portfolio & Profile Site

Live Site: evantwidwell.info

Overview

This is a full-stack web application serving as my professional portfolio and profile site. The project demonstrates expertise in data engineering, web development, and modern software architecture through a real-world implementation using NYC taxi data.

🚀 Key Features

1. Content Management System

• Custom Django-based CMS with markdown support
• Admin interface for content creation and management
• Python-frontmatter integration for structured content
• Dynamic post rendering with SEO optimization

2. Data Engineering Pipeline

• Data Source: NYC Taxi & Limousine Commission public datasets (millions of records)
• Database Engine: DuckDB for high-performance analytics
• Data Format: Optimized Parquet files for fast querying
• Caching Strategy: LRU cache implementation for API response optimization
• API Design: RESTful endpoints using Django REST Framework

3. Interactive Data Visualization

• Frontend Framework: React 19 with TypeScript
• Build Tool: Vite for fast development and optimized builds
• Mapping: Interactive Leaflet maps showing taxi zone data
• Charts: Recharts for statistical visualizations
• UI Framework: Tailwind CSS for responsive design
• Features: Date range filtering, real-time statistics, geographic data visualization

4. Professional Resume & About Page

• Comprehensive resume with work experience and skills
• Professional presentation of background and expertise

🛠 Technical Stack

Backend

• Framework: Django 4.2+
• API: Django REST Framework
• Database: DuckDB (analytical), PostgreSQL (application data)
• Data Processing: Pandas, PyArrow
• Deployment: Gunicorn, WhiteNoise for static files
• CORS: Django-CORS-Headers for cross-origin requests

Frontend

• Framework: React 19 with TypeScript
• Build System: Vite
• State Management: React Hooks
• HTTP Client: Axios
• Styling: Tailwind CSS v4
• Maps: Leaflet with React-Leaflet
• Charts: Recharts
• Date Handling: date-fns

Infrastructure

• Containerization: Docker with multi-stage builds
• Orchestration: Docker Compose
• Process Management: Makefile for development workflow
• Code Quality: ESLint, TypeScript strict mode

📊 Data Engineering Highlights

• Scale: Processing millions of NYC taxi trip records
• Performance: Sub-second query responses through DuckDB optimization
• Architecture: Separation of analytical (DuckDB) and transactional (PostgreSQL) workloads
• Efficiency: Parquet columnar storage for 10x faster analytics vs traditional formats
• Caching: Intelligent cache warming and LRU eviction strategies

🏗 Architecture

Frontend (React/TypeScript)
    ↓ HTTP/REST
Django REST API
    ↓ SQL
DuckDB + PostgreSQL
    ↓
Parquet Files

🚀 Getting Started

# Clone the repository
git clone https://github.com/evantwidwell/new_profile.git
cd new_profile

# Or use the Makefile
make up