Skills

This site, my first project using React/Next.js, serves as a comprehensive wiki for the Frieren TCG. It features detailed information about the game, including lists of cards and characters, strategies, and more.

Cardinal is my third attempt at creating a bot that works well for servers of various sizes. My first two attempts failed due to inadequate planning and practices. However, after about five months of continuous development, I have finally reached a point where I am satisfied with Cardinal. It has been verified, marking a significant achievement in my portfolio.

This project implements a Retrieval-Augmented Generation (RAG) system to allow users to query information about characters from the Frieren: Beyond Journey's End series based on parsed wiki data. It showcases a full RAG pipeline, including data loading, chunking, embedding using large language models (OpenAI/Google), and vector storage (with support for in-memory and ChromaDB). The system retrieves relevant context to ground responses generated by a separate LLM. It features both a command-line interface and a streaming web API with a simple static frontend, demonstrating practical application of AI/NLP concepts and backend development.

Gir was a self-funded Discord project that a friend and I worked on for quite a while. Our purpose was to create a bot which contained most of the most requested commands and systems that Discord users have been asking for. This bot first started out as a way for me to test certain commands and events, and it soon evolved into a full-fledged bot.