Machine learning engineer focused on reinforcement learning, NLP, multi-task learning, and post-training.
Website · LinkedIn · Writing · X · Email
I'm the sole ML engineer on the four-person founding team at a stealth startup, responsible for the machine learning work from experimentation and evaluation through implementation.
I also founded LiminalML, a SaaS company I design, build, and run solo. I make software for learning and explaining machine learning:
- Study — work through ML concepts with intuition, maths, code, and recall practice.
- Studio — create and share interactive technical visuals for articles, slides, and documentation.
LexiMind began as my independent undergraduate research project. It includes a transformer implemented from scratch in PyTorch, with FLAN-T5 weight transfer and a shared encoder for summarisation, emotion classification, and topic classification. My current interests include shared representations, adaptation, and how to use a fixed post-training compute budget well.
The repository also contains a book-discovery app. Its content-based recommendations run separately from the research model. Ongoing experiments and their limitations are documented in the research notes.
| Project | What I built | Links |
|---|---|---|
| Pulse | A local-first macOS app and menu-bar dashboard for coding-agent usage: subscription allowances, reset times, token history, and estimated API costs. Built with Electron and Node.js. | Site · Code |
| SartheAtlas | A Le Mans car and circuit explorer with interactive engineering studies, exploded views, terrain, and elevation profiles. Built with React, TypeScript, Three.js, and MapLibre; the engineering models are illustrative studies. | Site · Code |
| PlayAxis | Sport and gaming events, team following, a training log, and community features, built with React and FastAPI. | Site · Code |
| F1 Race Predictor | Classical machine learning on Formula 1 qualifying data, with a Streamlit dashboard. | Code |
I write about model implementations, research decisions, and the ideas behind my work on LiminalML's Substack.
B.S. in Computer Science, mathematics minor, and data science certificate from Appalachian State University. Based in the Charlotte metro area.