ADS-C
Implementation and experiment drivers for inference-time antidistillation in classification.
ML security · Reproducibility
SOFTWARE / EXPERIMENTS
I also build things.
Tools that turn ideas into working systems, from reproducible research experiments to multi-backend machine-learning libraries.
Research systems
Fontus
A custom aerial-recharging environment and PPO experiments investigating energy constraints, reward shaping, and an idle-agent failure mode.
Multi-agent environment · PPO / RLlib · Reward shaping
Selected engineering
Teras
A multi-backend library for deep learning on tabular data, built on Keras 3. Architectures for prediction, imputation, and synthetic-data generation across JAX, PyTorch, and TensorFlow.
Keras · JAX · PyTorch · TensorFlow
Keras
Open-source contributions to multi-backend ML infrastructure, including Beta and Binomial random-sampling APIs and backend consistency fixes.
Open source · Multi-backend APIs
codriver
An experiment in compact, local code generation: a custom 270M-parameter model trained through knowledge distillation, with separate language and fill-in-the-middle code training stages.
JAX · Flax · Knowledge distillation
Earlier work
A small archive of student and personal projects.
- Prewordict — A word-cloud guessing game built from clusters of words in Medium articles. GitHub ↗
- Gamer Assistant — A CS50 project combining game recommendations, PC suggestions, and FPS prediction.
- PROLEAs — Undergraduate work on crowd detection and behavior analysis using computer vision.
- FPS Predictor — An early machine-learning project estimating game performance from hardware data.
- Tweets Cloud Bot — A Twitter bot that turned recent tweets into customizable word clouds. GitHub ↗