Hi, I'm

Md. Fahim Sultan

PhD Student | Researcher — AI & Security

AI-security • LLMs • Machine Learning • Bioinformatics • Health Informatics

Oakland University
Rochester Hills, Michigan, USA

About

Md. Fahim Sultan is a Ph.D. student in Computer Science and Engineering at Oakland University, where he specializes in optimizing large pretrained language models (LLMs) for security-critical and low-resource environments. His research centers on advanced parameter-reduction techniques, adaptive fine-tuning strategies for domain-specific LLMs, and the development of secure, reliable AI frameworks for software engineering and cybersecurity applications. He has successfully completed a NASA MSGC–funded project on AI- driven vulnerability detection in aerospace systems, contributing novel methods for building robust, interpretable, and domain-adaptable AI solutions.
In addition to his primary research focus, Fahim has expanded his methodological contributions to the healthcare domain, where he has explored innovative approaches for developing efficient, context-aware medical AI models. His work emphasizes improving model generalizability, data-efficiency, and clinical reliability in low-resource healthcare settings, highlighting his versatility as an AI researcher. Driven by continuous learning and a passion for methodological innovation, he aims to advance scalable, trustworthy, and high-impact AI systems that address real-world challenges across security, aerospace, and healthcare domains.

“My last name ‘Sultan,’ meaning leadership in Arabic, reminds me to lead with purpose, humility, and continuous learning.”

Research Interests

AI Security

Adversarial robustness, secure model optimization, and safety-aligned LLM inference.

Large Language Models

PEFT, domain adaptation, and structure-aware optimization for code and reasoning tasks.

Software Intelligence

LLM-based vulnerability analysis, program reasoning, and neuro-symbolic modeling.

Biomedical Informatics

Clinical NLP, sequence modeling, and AI model deplyment for health systems.

Graph Learning

Graph-structured representations for biological sequences and security analysis.

LLM Quantization

Efficient compression and deployment of large language models through precision reduction, calibration, and hardware-aware optimization.

Embedding Architectures

Design and analysis of domain-specialized representation learning methods, including contrastive, graph-based, and transformer-driven embedding models.

Skills

Grant Writing: NIH (R01) Grants, NSF Grants, NASA MSGC Funded Projects
Programming & Frameworks: Python, C, R, PHP, PyTorch, TensorFlow, Keras, Scikit-learn, NumPy, Pandas
AI/ML Expertise: LLM, DL, ML, NLP, GNNs, Meta-learning, Ensemble Modeling
Data Science & Analytics: Exploratory/Statistical Analysis, Bioinformatics, Sequence Modeling, Gene Expression Analysis; Weka, HPC, iLearnPlus, iFeatureOmega
Security & Deployment: Docker, AWS, Joern; HTML/CSS, MySQL
Visualization & Reporting: Matplotlib/Plotly, Seaborn, Canva, Lucidchart/Draw.io
Research & Documentation: LaTeX, MS Office, EndNote, Scientific Writing & Experimentation

Experience

Graduate Research Assistant
Security & AI Research Lab Jan 2025 – Present

Teaching Assistant
Introduction to Computers and Programming with Excel, Dept. of CSE | Sep 2025 – Dec 2025

Research Assistant
Health Informatics & Research Lab (HIRL) | Jan 2023 – Feb 2024

News & Updates

Certificates

2025

Graduate Student Teacher Training — Oakland University, USA.

2025

Conference Speaker — 14th International Conference DATA 2025, Bilbao, Spain.

2024

Participation in AI Celebration Project Showcase — Daffodil International University, Bangladesh.

2022

Participation in 3rd ELC Economical Project Contest — Universitas Islam Sultan Agung, Indonesia.

2020

Advances in Computational Intelligence in Algorithms & Computing — IJCACI

2019

Academic Excellence Award — Dhaka Metropolitan Police (DMP), Bangladesh.

2017

Advance Certificate Course in Computer Application — NIT, Bangladesh.

Services (Reviewer)

  • Journal name: Journal of Cheminformatics, Springer Nature, Article Type: Full length article, Review on: 13 Dec. 2025.
  • Journal name: Diabetology & Metabolic Syndrome, Springer Nature, Article Type: Full length, Review on: 08 Nov. 2025.
  • Conference name: IEEE BigData 2025, Article Type: Short length article, Review on: 29 Oct. 2025.
  • Conference name: IEEE BigData 2025, Article Type: Short length article, Review on: 17 Oct. 2025.
  • Journal name: Scientific Reports, Article Type: Original Research, Review on: 27 Aug 2025.
  • Journal name: Molecular Diversity, Article Type: Original Research, Review on: 21 Aug. 2025.

Contact & Connect