CS teacher, CS grad student, and builder focused on ML systems, Kaggle competitions, game AI, and local evaluation tooling.
I like projects where the output is measurable: benchmark runs, agents, simulations, dashboards, model pipelines, or tools that make experimentation faster.
- Kaggle AI competitions and local benchmark harnesses
- Pokemon TCG AI Battle evaluation tooling
- Game AI / Roblox AI playtest systems
- Cybersecurity benchmark literature survey for LLMs
- GPU-backed local AI apps and inference workflows
| Project | What it does | Stack |
|---|---|---|
ptcg-meta-bench |
Local meta-weighted benchmark harness for Pokemon TCG AI agents | Python, Kaggle SDK, simulation eval |
GolfWebGUI |
Golf analytics / web tooling project | Python, web UI |
BirdCLEF 2026 |
Audio ML competition notebooks and reproducibility work | Python, Kaggle, ML |
Roblox FPS AI Lab |
Server-authoritative FPS framework with bot playtesting | Lua, Roblox Studio, MCP |
Cyber Benchmark Survey |
Literature survey on what LLM cyber benchmarks actually measure | Research, benchmark analysis |
Local AI Tooling |
Experiments with vLLM, local models, agents, and GPU workflows | Python, vLLM, Linux |
- Evaluation harnesses
- AI agents
- Game simulation tools
- ML competition pipelines
- Research tooling
- Local-first AI apps
- Kaggle:
add your Kaggle link - Hugging Face:
add if useful - Portfolio / blog:
add later