AI Engineer · Builder · Computer Science Student
I build practical software, explore AI systems, and contribute to open source.
|
|
AI engineering · developer tooling · automation · databases · open source
I like turning ideas into working software, from AI applications and database tooling to developer utilities, automation, and interactive systems.
My current focus includes:
- AI & intelligent applications — Python, Azure AI, LangChain
- Developer tooling — CLI applications, validation, testing, automation, CI/CD
- Systems & data — databases, APIs, input validation, defensive programming
- Interactive software — real time visualization and computer vision experiments
- Cloud & deployment — Azure, Google Cloud, Netlify, Vercel
I work on existing codebases as well as my own projects: understanding unfamiliar code, fixing concrete problems, adding regression coverage, and responding to maintainer feedback.
| Contribution | What I worked on |
|---|---|
| Soup #1243 · merged | Fixed Transformers 5 BatchEncoding handling in sequence distillation and added regression coverage |
| Soup #1053 · merged | Fixed Rich markup handling for user-controlled model names and added regression coverage |
| Provena #182 · merged | Added regression coverage for the top-level CLI commands exposed in help output |
| pycubrid #378 · open | Added strict positive-integer validation for Cursor.arraysize across sync and async cursors |
| cubrid-mcp-server #1 · open | Fixed SQL comment handling in audit categorization with regression coverage |
The goal is simple: find a real problem, understand the existing implementation, make the smallest reliable change, and prove it with tests.
A small SQLite style database engine built from scratch in C, covering SQL parsing, table operations, B-tree storage, paging, persistence, testing, benchmarking, and sanitizers.
A lightweight search engine built from scratch in Python with tokenization, an inverted index, TF-IDF ranking, Boolean queries, phrase search, JSON persistence, CLI support, tests, and CI.
A Python DB-API 2.0 driver for CUBRID, with synchronous and asynchronous interfaces, SSL/TLS support, and extensive testing.
An MCP server for working with CUBRID databases, with an emphasis on SQL safety, audit behavior, and reliable database tooling.
Understand → change → test → verify → document.
I care about software that is understandable, reproducible, and useful, not just code that happens to run once.
The contribution Snake is refreshed automatically from GitHub's public contribution data.