Machine Learning Engineer. I build and evaluate models on neural and clinical signal data.
MS Computer Science, Machine Learning specialization — Georgia Tech (in progress). Software developer by day: Python and SQL pipelines on Databricks.
- EEG decoding — deep learning on neural signals that generalizes across subjects, not just across epochs
- Reinforcement learning — sample efficiency and stability; recently implemented TD3 from scratch in PyTorch
- Algorithmic foundations — why a method works, not only that it scored well
| Project | What it is |
|---|---|
| liminal-eeg | Predicting experience intensity from EEG. MNE + PyTorch pipeline over 34 subjects; EEGNet and a temporal Transformer benchmarked against a published SVM baseline under leave-one-subject-out CV. |
| Prostate-Segmentation | 3D prostate MRI segmentation with nnU-Net v2 on Medical Segmentation Decathlon Task05. Done as a CS research assistant. |
| intrusion-detection-system-using-decision-trees | R2L attack detection on KDD Cup 1999 and NSL-KDD. Combining the datasets reached 99.93% accuracy at a 0.08% false-alarm rate. Poster |
| Bill_Coleman_Project | AI for early detection of small cell lung cancer. In progress. |
I think about machine consciousness and where ML theory, neuroscience, and physics overlap. That intersection drives most of what I build.