Portfolio

A Firebase-powered social discovery platform where communities come alive through real-time connections. Think Facebook meets Tinder for building genuine friendships and communities - users can create rich content with posts and videos, discover people through swipe-based matching, join interest-based groups, attend local events, and engage in real-time messaging. The web app serves as the in-depth engagement hub for content creation and community management, while the mobile app focuses on real-time interactions and location-based discovery. Built for real customers with dual Firebase environments (staging for development, production for live users) and deployed on Vercel serving thousands of active community members.

A full-stack luxury experience platform that transforms digital gifting into emotionally-resonant moments. Imagine Instagram for memories meets AI-powered travel planner meets corporate gifting suite - users create beautiful memory journals with photos and videos to capture life's meaningful moments, get personalized travel recommendations from Clevis AI (powered by Google Gemini) through an interactive chat-based itinerary builder with Google Maps integration, and companies can manage bulk gift campaigns with employee roster uploads and analytics dashboards. Built with Next.js 15 (React 19, TypeScript) frontend and Express backend with Prisma ORM connecting to PostgreSQL, featuring Cloudinary media storage, Amazon SES emails, JWT authentication, and role-based access control. A complete ecosystem for personal memories, dream vacations, and corporate appreciation.

A sophisticated mineral rights analysis platform that transforms oil & gas property evaluation through interactive mapping and financial modeling. Imagine having the power to analyze entire states of oil & gas data - visualizing thousands of wells, townships, and land sections on an interactive MapLibre GL JS map with heat maps showing peak production, rig density, and section values, while running complex financial calculations including hyperbolic decline curves for production forecasting, NPV calculations with multiple discount rate scenarios, and economic modeling that accounts for royalty rates, deduction rates, operating expenses, and capital costs. Built as a pnpm monorepo with Next.js 14 (React 18, TypeScript, Mantine v7) frontend featuring Zustand state management, TanStack Query for data fetching, and Recharts/Chart.js/ECharts for visualizations paired with a Python FastAPI backend using SQLAlchemy, Pandas, and NumPy to crunch the numbers. The platform processes data across a 5-level geographic hierarchy (States → Counties → Townships → Sections → Wells), generates PDP (Producing Developed Petroleum) and PUD (Proved Undeveloped) forecasts, offers multi-scenario analysis (low/expected/high), creates PDF reports for premium users, and integrates with PostgreSQL via Supabase for real-time data access. From drilling to dollars, every decision backed by data.

High-performance web scraper for MyProtein products with JavaScript rendering capabilities. Extracts comprehensive product data including flavors, sizes, prices, nutritional information, and product variations using Scrapy and Selenium WebDriver for dynamic content handling.

A complete full-stack PC rental ecosystem for internet cafes that combines a powerful web admin dashboard with Electron desktop clients deployed on every rental machine. Imagine a cyber cafe management system where admins can monitor all PCs in real-time through beautiful analytics dashboards with charts showing revenue, session statistics, and PC utilization - while customers interact with sleek kiosk interfaces on each rental computer to start/end sessions and make payments. Built with Next.js 15 (React 19, TypeScript) for the web dashboard and Electron for the desktop clients, all connected to Supabase PostgreSQL with real-time subscriptions that instantly sync data across all devices. Features Zustand stores for client-side state management, Tremor & Recharts for stunning analytics visualizations, bcryptjs for secure password hashing, and automated builds for individual PC deployments. The system tracks every second of usage, processes payments accurately, generates comprehensive reports, and provides a seamless experience from customer login to admin oversight.

A sophisticated Google My Business scraper engineered for maximum efficiency and data quality. Features 13 parallel Chrome browsers with anti-detection mechanisms, processing 25 cities across British Columbia with 57 business categories. Built for lead generation with intelligent website filtering and real-time data validation.

A comprehensive full-stack restaurant ordering system with customer-facing menus, kiosk mode, real-time order tracking, and admin dashboard. Features three interfaces: customer menu, kiosk ordering, and admin portal for managing products, orders, and revenue analytics with receipt printing capabilities.

A sophisticated Python package providing AI-powered conversational interfaces for oil & gas data analysis. Built for the platform, it implements a two-tier bot architecture (Basic and Premium) with comprehensive database integration, web search capabilities, and advanced analytics features.

A sophisticated multi-state regulatory data extraction and management system specifically designed for oil & gas spacing orders and operator information. The platform integrates directly with the PostgreSQL database to provide real-time regulatory data availability across multiple states with ultra-fast database operations achieving 7,799+ records/sec.

A sophisticated oil & gas data processing system designed for targeted investment analysis across multiple US states. This system processes large-scale energy data (~23,782 wells) to support investment decision-making in the oil & gas sector with nightly updates and zero-downtime deployments.

A sophisticated Model Context Protocol (MCP) server designed for safe PostgreSQL database interactions with Supabase. Provides both read and controlled write access while maintaining security and performance safeguards specifically for PC rental management systems.

A sophisticated, production-ready Model Context Protocol server specifically designed for oil & gas industry data management with advanced PostgreSQL integration on AWS RDS. Features comprehensive safety controls, geospatial operations, and industry-specific tools for petroleum data analysis.

A comprehensive system for managing AI interactions through memory management, lessons learned tracking, and dual-mode operation (Plan/Agent). Ensures consistent, high-quality development while maintaining detailed project documentation and knowledge retention.

Enhanced knowledge graph memory server that lets Claude remember information about users across chats and learn from past errors through a lesson system. Built on the MCP protocol for AI memory management.