Portrait of Prity Rani Das

Software Engineer & AI Researcher

Prity Rani Das

Software Engineer at BRAC IT Services Limited. AI researcher specializing in hybrid deep learning for medical imaging. Ranked 5th in the Department of Software Engineering at NSTU with a CGPA of 3.68/4.00.

Scroll down

About Me

I am a Software Engineer at BRAC IT Services Limited and a passionate AI researcher with a deep commitment to leveraging technology for meaningful societal impact. My professional journey bridges the precision of enterprise software development with the innovation of cutting-edge deep learning research in medical imaging.

I completed my Bachelor of Science in Software Engineering from Noakhali Science and Technology University (NSTU), graduating with a CGPA of 3.68/4.00 and earning the 5th position in my department. My academic foundation was built upon consistent excellence, achieving a perfect GPA of 5.00/5.00 in both my SSC and HSC examinations.

What sets my trajectory apart is my ability to combine disciplined academic growth with independent research initiative. I produced a state-of-the-art hybrid deep learning model for polyp segmentation as a sole author, achieving an IoU score of 0.966. This accomplishment deepened my resolve to pursue advanced research at the intersection of AI and healthcare on a global stage.

Academic Highlights

  • BSc in Software Engineering

    NSTU | CGPA 3.68/4.00 | Rank: 5th

  • HSC Examination

    Dr. Mahbubur Rahman Mollah College | GPA 5.00/5.00

  • SSC Examination

    Hazi Ekhlas Uddin Bhuiyan High School | GPA 5.00/5.00

  • Research Paper

    Sole-authored paper on hybrid U-Net + ViT for medical imaging

  • Leadership

    Former Child Reporter | Early leadership & social engagement

Research

Sole Author

A Novel Hybrid Deep Learning Model Using U-Net and Vision Transformer for Polyp Segmentation in Medical Imaging

Preprint

Innovation

Designed a hybrid architecture integrating CNN-based U-Net with Vision Transformers (ViT) to capture both local spatial features and global contextual dependencies for precise polyp segmentation.

Key Result

Achieved a state-of-the-art Intersection over Union (IoU) score of 0.966 on the multicenter PolypDB dataset, demonstrating exceptional segmentation accuracy.

Motivation

Conducted independently as a sole-authored effort, driven by a deep commitment to advancing AI-assisted diagnostics and measurable research excellence.

Deep LearningU-NetVision TransformerMedical ImagingPolyp SegmentationCNNPolypDB

Experience

Jul 2024 — Present

Software Engineer

BRAC IT Services Limited

Primarily responsible for frontend development, while maintaining strong backend expertise in .NET architecture and system design. Contributing to enterprise-grade applications that serve millions of users through BRAC's extensive development and microfinance ecosystem.

Frontend Development.NETSystem DesignEnterprise Software

Apr — Jul 2024

Software Engineering Intern

BRAC IT Services Limited

Completed a rigorous 3-month intensive internship, demonstrating strong technical capability and adaptability that led to a direct transition into a full-time engineering role. Gained hands-on experience with production-level systems and enterprise development workflows.

Full-Stack DevelopmentProduction SystemsAgile

Projects

Social Impact

NSTU Medical Center Automation

Led the digital transformation of the university medical center's manual operations into a fully automated system. Significantly improved healthcare efficiency, patient record management, and service delivery across the entire NSTU campus.

Healthcare ITSystem DesignAutomationSocial Impact
Deep Learning

Skin Disease Classification

Developed a deep learning model leveraging transfer learning techniques for accurate classification of dermatological conditions. Applied convolutional neural network architectures to real-world medical image datasets.

CNNTransfer LearningImage ClassificationMedical AI
Image Processing

Leaf Feature Extraction

Built an advanced image processing pipeline using contour-based representation methods for automated extraction and analysis of leaf morphological features, supporting agricultural research.

Image ProcessingContour DetectionFeature ExtractionOpenCV

Contact

Interested in collaboration, research opportunities, or just want to connect?

I am actively seeking international graduate research opportunities in AI and medical imaging. I welcome inquiries from academic institutions, research supervisors, and fellow researchers.