Problem/Motivation: Learning advanced Software Design Patterns often lacks practical, real-world frontend implementation examples. Solution/Benefit: Developed a portfolio website using TypeScript and Next.js, explicitly implementing GoF patterns (Factory, Builder, Visitor) within the architecture. This resulted in a highly maintainable codebase and served as a tangible proof-of-concept for clean code engineering.

Problem/Motivation: Translating comics manually is highly inefficient due to the complexity of extracting text from images. Solution/Benefit: Engineered an end-to-end Python pipeline leveraging digital image processing and Optical Character Recognition (OCR). Integrated a Human-in-the-Loop (HITL) workflow to ensure high contextual accuracy and quality control in English-to-Thai translations.

Problem/Motivation: Using a single, static resume for different roles reduces the chance of matching company-specific requirements.Solution/Benefit: Designed a normalized relational database to securely and flexibly manage skills and experiences. Built a Dynamic Resume generator that outputs tailored documents mapped to specific Job Descriptions, laying the architectural groundwork for future LLM-driven candidate-job matching.

Problem/Motivation: Manually managing schedules and checking for appointment overlaps is time-consuming and error-prone. Solution/Benefit: Developed a Python-based AI agent utilizing the Model Context Protocol (MCP) to interpret natural language commands. Integrated with the Google Calendar API to automate scheduling and perform real-time overlap detection, significantly reducing manual calendar management.

Problem/Motivation: Setting up new software project structures involves repetitive and time-consuming manual configurations. Solution/Benefit: Built a command-line interface (CLI) to automate project bootstrapping (Selection & Contribute features). Engineered a highly flexible architecture designed to support future LLM integration, enabling developers to scaffold custom projects using simple natural language prompts.

Problem/Motivation: Networking events often suffer from low engagement and awkward initial interactions. Solution/Benefit: Developed a `Phygital` platform merging physical and online participation. Applied AI to dynamically generate and randomize domain-specific questions (e.g., Computer Science, Data Science), effectively breaking the ice and fostering meaningful professional connections.

A massive e-commerce ecosystem. Includes user authentication, product management, shopping cart, and payment gateway integration.

Admin panel for sellers.

A local-first interactive Todo mini-app built into the portfolio.