🎓 MS in Computer Science Candidate – University of Minnesota Duluth (2026–2028)
🔬 Applied AI Researcher | 16+ years experience | 6+ publications
💼 Additional Director, ICT Cell – University of Dhaka | CTO – Phoenix Software Ltd
I am an applied AI researcher with 16+ years of experience driving large‑scale digital transformation and data‑centric innovation at the University of Dhaka. My research bridges healthcare AI, federated learning, deep reinforcement learning, and computer vision to build trustworthy, explainable, and privacy‑preserving systems.
Currently pursuing my MS in CS at UMN Duluth to deepen theoretical foundations and contribute impactful research at the intersection of AI and clinical decision support.
| Repository | Focus |
|---|---|
federated-explainable-ai-chronic-disease |
Multimodal deep RL + federated learning for early chronic disease prediction (under review) |
yolov8-toll-anomaly-detection |
Real‑time YOLOv8 + geospatial anomaly detection in road transport networks |
lung-colon-cancer-cnn |
CNN‑based classification (93% accuracy) from histopathological images |
federated-android-malware-detection |
Privacy‑preserving federated deep learning (IEEE QPAIN 2026) |
graph-neural-network-anomaly |
GNN for real‑time intrusion detection in large‑scale networks |
lstm-stock-prediction-fullstack |
LSTM + Django REST + ReactJS full‑stack prediction system |
- Federated, Explainable, and Privacy‑Preserving AI for Early Detection of Chronic Diseases – Under review (2026)
- Explainable Multimodal AI Framework with RL for Post‑Surgical Clinical Decision Support – Under review (2026)
- Secure Federated Deep Learning for Android Malware Detection – IEEE QPAIN 2026 (conference)
- Statistical Reliability of ML Intrusion Detection for High‑Traffic Networks – Control and Decision, vol. 40, no. 10, 2025
- Quantitative Study of Neural Network‑Based Anomaly Detection – Control and Decision, 2025
| Category | Technologies |
|---|---|
| Languages | Python, PHP, JavaScript/TypeScript, C/C++, SQL |
| AI/ML Frameworks | TensorFlow, PyTorch, Keras, scikit‑learn, OpenCV, Pandas/NumPy |
| Specialties | Deep Learning, Reinforcement Learning (DQN, A2C, DDPG), Federated Learning, Graph Neural Networks |
| Backend & DevOps | Django, FastAPI, Laravel, Docker, Kubernetes, AWS/GCP/Azure, CI/CD |
| Databases | PostgreSQL, MongoDB, Elasticsearch, MariaDB |
- Academic email:
mostak@duluth.umn.edu(preferred) - GitHub: github.com/mostakphoenixsoftbd
⚡ Standardised test scores: GRE 329 (Q170/V159/AWA5.0) | TOEFL 107
⚡ Fun fact: I once debugged a production ML pipeline for 3 hours – the issue was a missing reshape().