I build AI systems and developer tools.
CS @ Beaconhouse National University | AI Researcher @ SCIT CAIR | NeurIPS '24 Workshop Author
| Project | About | Stack |
|---|---|---|
| VoiceFlow | Local voice dictation on faster-whisper. Runs on your GPU, nothing leaves your machine. | TypeScript faster-whisper Pyloid |
| Vision-Dissect | Tears apart 289+ layers of CV models to benchmark depth estimation, object detection, and segmentation side by side. | PyTorch ONNX Transformers |
| Anonator | Anonymizes faces in video with multiple detection backends. HIPAA compliance mode for medical footage. | YOLOv8 OpenCV CUDA |
Mapping Air Pollution Sources with Sequential Transformer Chaining NeurIPS 2024 Climate Change AI Workshop
We chained Vision Transformers with CLIP filtering to find pollution sources in Sentinel-2 satellite imagery. Filtered 600K+ images down to the ~1% that actually contained sources.