
QWOP with PPO
A PyTorch PPO agent plays the real QWOP browser game through Playwright. Six evaluation races, six 100 m finishes. Grace was not part of the reward function.
AI/ML
Text, images and audio, with review built in
A Rust + React labeling platform with configurable schemas, multi-annotator agreement and optional model-assisted pre-labeling. One Docker image to self-host.

A Rust + React labeling platform with configurable schemas, multi-annotator agreement and optional model-assisted pre-labeling. One Docker image to self-host. Rewritten from the original Flask application in 2026, keeping annotation, review and export in one workspace.

A PyTorch PPO agent plays the real QWOP browser game through Playwright. Six evaluation races, six 100 m finishes. Grace was not part of the reward function.

An end-to-end Nipype pipeline with FSL FAST, K-means and ANTs Atropos segmentation, tissue volumes, bias-correction metrics and an interactive 3D brain.

Compare EasyOCR, RapidOCR and PaddleOCR on Traditional Chinese street signs, with text boxes, timings and a small public-domain benchmark.