CV

Mahbuba Tasmin
PhD Candidate in Computer Science (Computational Biology & Machine Learning), University of Massachusetts Amherst
· linkedin.com/in/mahbuba-tasmin · github.com/Tasmin153 · Google Scholar

Education

• University of Massachusetts Amherst, Amherst, MA
  Ph.D. Candidate in Computer Science (Advisor: Prof. Anna Green), Expected May 2027
  • Research Interests: Resistance Forecasting, Explainable ML, Protein Language Models, Structural Bioinformatics
  • Awards: Sudha and Rajesh Jha Scholarship (2023)
  • Relevant Coursework: Advanced Algorithms, Machine Learning, Neural Networks, Computational Biology, AI, Information Assurance, Computer Architecture, Teaching for Tomorrow’s Faculty, ML for Biological Sequence Data

• University of Massachusetts Amherst, Amherst, MA
  M.S. in Computer Science • Thesis aligned with Ph.D. research on antibiotic resistance modeling

• North South University, Dhaka, Bangladesh
  B.S. in Computer Science and Engineering (Summa Cum Laude) • Concentration: Artificial Intelligence and Algorithms

Research Experience

• Graduate Research Assistant, SAGE Lab, University of Massachusetts Amherst — Amherst, MA
  Sep. 2023 – Present
  • Lead researcher on BIG-TB, a multimodal benchmark of ~17K M. tuberculosis isolates for resistance prediction across 11 WHO-priority drugs
  • Designed sequence- and structure-aware models (CNNs, Transformers, fused ridge baselines) integrating DNA/protein features and ESM embeddings
  • Built multi-species augmentation pipelines (UniProt, InterPro) to improve protein-level generalization via evolutionary data
  • Performed explainability studies with SHAP and causal variant recovery (recall@k) versus WHO 2023 catalog
  • Led Forecasting Antibiotic Resistance Using Biophysics and ML, fusing protein thermostability metrics with machine learning for variant assessment
  • Collaborated with Harvard DBMI (Farhat Lab) on reproducible multi-gene modeling and benchmarking
  • Formulated a fused ridge regression framework with convex regularization, fusion penalties, and accelerated gradient descent (momentum, clipping, Nesterov)

Industry Experience

• AI Engineer, NITEX Solutions Ltd. — Dhaka, Bangladesh
  2022
  • Implemented Detectron2-based instance segmentation for product identification and OCR automation tools
  • Built fashion trend moodboards combining NLP and computer vision pipelines for workflow automation

• Software Engineer (AI & ML), M2SYS Technology — Dhaka, Bangladesh
  2020 – 2022
  • Developed image spoofing detection and contextual recommendation systems using ML and NLP techniques
  • Automated backend workflows with Camunda and deployed production ML models across distributed systems

Selected Publications

• Green, A. G., Tasmin, M., Vargas, R., Farhat, M. R. “The structural context of mutations in proteins predicts their effect on antibiotic resistance.” Submitted to eLife. Preprint: bioRxiv 2025.09.23.676583 (2025)
• Tasmin, M., Green, A. “Beyond Sequence-only Models: Leveraging Structural Constraints for Antibiotic Resistance Prediction in Sparse Genomic Datasets.” ICLR 2025 MLGenX Workshop (2025)
• Yang, Z., Yao, Z., Tasmin, M. et al. “Unveiling GPT-4V’s hidden challenges behind high accuracy on USMLE questions.” Journal of Medical Internet Research (2025)
• Tasmin, M. “Protein Structure-Informed Regularized Linear Model Outperforms ESM for Predicting Antibiotic Resistance.” Program in Quantitative Genomics (PQG), Harvard University, 2024 — Poster presentation demonstrating integration of 3D structural features with machine learning models

Talks and Presentations

• BIG-TB: A Benchmark Dataset for Genomic Resistance Prediction and Interpretability in Mycobacterium tuberculosis — Machine Learning for Computational Biology (MLCB) Workshop, 2025 · Spotlight talk highlighting dataset design and explainability analyses
• Protein Structure-Informed Regularized Linear Model Outperforms ESM for Predicting Antibiotic Resistance — Program in Quantitative Genomics Conference (PQG), Harvard University, 2024 · Poster presentation on 3D structural integration with ML models

Teaching Experience

• Head Teaching Assistant, CompSci 520: Theory and Implementation of Advanced Software Engineering, UMass Amherst
  Fall 2025 – Spring 2025 (academic year)
  • Led 140+ students, managed TA team, coordinated logistics, and maintained GitHub Classroom and Gradescope automation
• Course Developer Assistant, CompSci 520: Theory and Implementation of Advanced Software Engineering, UMass Amherst
  Summer 2023
  • Revamped course structure, labs, and assignments with automated grading workflows

Honors and Awards

• CRA-WP Grad Cohort for Women (2023), San Francisco, USA

Technical Skills

• Tools & Languages: Python, Bash, R, LaTeX, pandas, NumPy, matplotlib, Git, Docker, Linux, SLURM
• ML/AI: PyTorch, scikit-learn, XGBoost, CNNs, Transformers, Random Forests, SHAP
• Bioinformatics: UniProt, InterPro, Rosetta, AAIndex, Protein Structure analysis, VEP/ANNOVAR

Leadership & Mentorship

• PhD Graduate Representative, Student Representative in Faculty Senate, College of Computer & Information Sciences, UMass Amherst — Fall 2025
• Shakir Sahibul, M.S. Student, UMass Amherst — Fall 2024 · Supervised Transformer-based resistance prediction project
• Suqi Hong, M.S. Student, UMass Amherst — Fall 2025 · Supervising EvoAug-based protein resistance prediction project