- Vision models: representations, architecture, applications
⚡ Exploring. Building. Breaking. Learning faster.
⚡ Exploring. Building. Breaking. Learning faster.
A failure-aware harness for reproducible quantitative-strategy research, with typed agent workflows and inspectable experiment memory.
A lightweight multimodal brain-encoding research prototype distilled from TRIBE v2 for predicting fMRI responses to naturalistic video.
Python 2
A Kafka-driven post-trade reconciliation engine with audit trails, metrics, dashboards, and mismatch reporting.
Deploy PyTorch vision models as fast, memory-safe Rust binaries; convert to ONNX/TorchScript, quantize, benchmark, and package into a native, REST, or WASM target with one CLI.
Python 3
Evaluating and enhancing CLIP’s zero-shot and fine-tuned performance for melanoma detection for improved accuracy.
A toolkit for unseen object segmentation tasks