BrainT_XAI_RAG
↗Explainable Vision AI pipeline for brain tumor detection using EfficientNetB3, Grad-CAM and a Visual RAG layer. Achieved ~91.5% accuracy.
Computer Science and Engineering graduate specializing in Computer Vision, Artificial Intelligence, Machine Learning, and Deep Learning. Former paid Intern at the ICT Cell, Bridges Division, Government of the People’s Republic of Bangladesh. Erasmus+ scholar at Istanbul Kültür University. Strong background in AI-driven application development and secure system design.
Explainable Vision AI pipeline for brain tumor detection using EfficientNetB3, Grad-CAM and a Visual RAG layer. Achieved ~91.5% accuracy.
Semantic segmentation for dental caries using FastViT; 92.5% pixel accuracy and 79.25% mIoU.
Interpreted a breast cancer classification model using SHAP and produced feature-importance visualizations.
Streamlit + ChromaDB RAG app for citation-aware QA and similarity search over documents.
BSc in Computer Science & Engineering
Daffodil International University (June 2020 – Sept 2024). Final grade: 3.53 / 4.00.
LRMC-DeepLabV3+: Multiclass Leaf Disease Semantic Segmentation — proposed an improved DeepLabV3+ model evaluated on six plant leaf classes; reported 97.34% accuracy and 93.47% mIoU. Roles: methodology, coding, dataset curation, and first authorship.
Paid Intern (ICT) — Government Bridges Division (31 Mar 2025 – 30 Jun 2025). Supported ICT operations and digital infrastructure at the Bridges Division.