Open to opportunities

Asif Ahemed

@asifahemed

I build machine learning systems and scalable React/Node.js applications.

I'm building multimodal deep learning systems at BRAC University, including a framework to predict Alzheimer’s Disease progression from longitudinal MRI datasets and a GNN-based DistilBERT system for multi-label music context prediction.

I've trained predictive models on 500K+ records, achieving 83% accuracy for traffic accident severity prediction while doubling recall for high-severity crashes. I also build full-stack products, including Disaster Helper, a React, Express, and MongoDB application for coordinating emergency relief logistics.

Experience

Work history, roles, and key accomplishments

Current

Research Assistant

Current

BRAC University

Jan 2026 - Present (8 months)

Spearheaded the architectural design of a multimodal deep learning framework to predict Alzheimer's Disease progression from Mild Cognitive Impairment using longitudinal MRI datasets. Prototyped a parameter-efficient 3D CNN integrated with a Multi-Head Self-Attention Transformer and engineered a robust data ingestion pipeline with Expected Calibration Error methods.

PyTorch Multi Head Attention Transformers Data Pipelines Deep Learning Multimodal Learning

Education

Degrees, certifications, and relevant coursework

BRAC University

Bachelor of Science, Computer Science

2023 -

Grade: 3.36/4.0

Activities and societies: GNN-Based BERT for Understanding Context from Music; Research on multimodal deep learning for Alzheimer's Disease progression.

Pursuing a Bachelor of Science in Computer Science with a focus on machine learning and deep learning. Currently maintaining a CGPA of 3.36/4.0.