wafer-inspection

Drift-Sense solves the Navigation-Error Recovery problem. By utilizing a hybrid architecture of classical vision (NCC) and a lightweight Siamese CNN ranker, it accurately matches high-resolution reference patterns against noisy, low-resolution SEM images without heavy transformers.

Sub-pixel die-site recovery for wafer inspection (Applied Materials PS02, SEMICON India) — classical CV grounded in SEM physics, CPU-only, rank-0 match at 0.43 px

Drift-Sense solves the Navigation-Error Recovery problem. By utilizing a hybrid architecture of classical vision (NCC) and a lightweight Siamese CNN ranker, it accurately matches high-resolution reference patterns against noisy, low-resolution SEM images without heavy transformers.

Distributed controller design for wafer inspection: fans out one invocation per frame-in-die position to compute nodes reading frames from a shared memory box, gathering one value per position per slice

Wafer edge defect detection system - C#/.NET 8/WPF/OpenCvSharp/ONNX with synthetic data and real-time visualization

Python simulation of semiconductor wafer defect inspection, optical detection, and defect-limited yield

After-work engineering notes connecting materials science, semiconductor measurements, process monitoring, failure localization, and the evidence limits of inspection.

NAFNet-SR (2.39M parameters) deep learning solution for semiconductor wafer image restoration and 2x super-resolution. Achieves 28.98 dB PSNR / 0.776 SSIM (+5.97 dB gain) with real-time GPU inference. KLA Hackathon 2026.

FastAPI and SQLModel reference backend for wafer-inspection sessions, runtime state, result aggregation, and synthetic Excel export.