Redefining what intelligent systems can do — from the lab to production.
Washikur Rahman is a Researcher and AI Engineer graduating with Highest Distinction from BRAC University (CGPA 3.88/4.00). Currently working as a graduate teaching assistant at USASK.
His work spans Graph Theory, Graph Visualization, computer vision for industrial automation — combining deep research foundations with production-grade engineering.
Modular pipeline for supervised learning — classification and regression across diverse datasets. Automated experiment tracking, hyperparameter optimization, and full reproducibility via MLflow.
ML-based computer vision for identifying defects in manufactured components with high precision. Deployed in an industrial automation pipeline, significantly reducing manual inspection effort.
Retrieval-Augmented Generation architecture for intelligent domain-specific query answering. Vector search and contextual response generation for enterprise knowledge retrieval.
Proposed a hybrid algorithm addressing color distortion, low contrast, and haze in underwater environments. Combined Discrete Wavelet Transform with CLAHE and bilateral filtering in a recursive framework. Validated with PSNR and SSIM metrics; demonstrated improved YOLO-based object detection accuracy.
Investigating decoding of visual stimuli from EEG signals by analyzing brain activity in the visual cortex. Applied signal processing and supervised learning to map EEG patterns to image features for brain-computer interfacing. Targeting visual reconstruction for cognitive neuroscience and assistive technologies.