Bio

Summary

I work on AI/ML that is reliable, interpretable, and reproducible, especially for biomedical and other sensitive datasets. My recent work spans multi-omics modeling (e.g., cancer experiments with transformer/GNN-based approaches), optimization-driven model selection (including hybrid quantum–classical methods such as QAOA and VQE in Qiskit), and privacy-preserving analytics (FHE/HE, secure aggregation, and post-quantum cryptography). I also build end-to-end healthcare analytics and clinical NLP pipelines, and I have applied modern NLP tooling to finance (e.g., sentiment/risk-oriented workflows). I’m comfortable with Python, PyTorch, Hugging Face, SQL, and Qiskit, and I prioritize clean, reproducible repos with pinned environments and clear run instructions. For a PhD, I’m interested in optimization for large models, robustness and privacy, and hardware-aware ML systems that run on real infrastructure—cloud/GPU, edge/IoT, and near-term/noisy quantum backends—across NLP, time-series, tabular, and multimodal data.

Experience

• Sep 2024 – Present Associate Researcher (Independent Contractor) — Merrimack College, North Andover, MA
  • Designed and evaluated 5+ cancer-model experiments (GNNs/transformers) on multi-omics (TCGA-BRCA, MIMIC), delivering +6 pts AUROC (0.94 vs 0.88) and +8% F1-score vs. classical PCA baselines.
  • Built 3+ hybrid quantum-classical prototypes in Qiskit (QAOA/VQE) for classification & optimization, improving model precision by +36 pts (78% vs 42%) in target prioritization and accuracy by +11 pts (89% vs 78%) in FinBERT HPO.
  • Produced 11+ literature reviews and 45+ figures/tables for publications; co-authored 9+ manuscripts (conference, journal, book chapter) and introduced LaTeX/Zotero templates that cut drafting/review cycles by 30%.
• May 2025 – Present Data Analyst (Contract) — Next Tech USA LLC, Jamaica, NY
  • Ingested, cleaned, and documented 15+ large healthcare datasets (50M+ rows), building EDA/feature pipelines that reduced data defects by 30% and shortened analysis cycle time by $5~hrs/week$.
  • Developed and validated 3 ML models for patient-outcome prediction and workflow optimization, improving AUC by +8 pts and reducing false alerts by 15%; shipped 2 models to staging.
  • Built 10+ stakeholder dashboards (Power BI/Tableau) with 25+ KPIs, driving adoption across 4 teams and cutting report prep time from 8 hrs to 30 mins per week via automated refresh and role-based views.
• May 2024 – April 2025 Data Science Intern (Quantum Computing Research, ML Applications, and Data Analytics) — Tekurai Inc., San Antonio, TX
  • Explored protein folding with GNNs and IBM Qiskit; built transformer/CNN pipelines for clinical text and audio.
  • Processed large healthcare datasets; delivered Tableau dashboards for operational insights.
  • Contributed to blockchain-enhanced, privacy-aware data workflows for drug discovery.
• Sep 2024 – Present Teaching Assistant — Merrimack College, MA
  • DSE 4900 (Capstone) – Fall 2024; Fall 2025: Mentored teams from proposal to final presentation; coordinated labs and weekly clinics; supervised projects/theses end-to-end (scoping, curation, modeling, reporting); instituted reproducible practice (clean repos, env pinning, MLflow) and rigorous evaluation (ablations, error analysis, calibration, leakage/imbalance).
  • DSE 4116/6116 Quantum Machine Learning – Fall 2025: Designed and delivered Qiskit labs (VQE, QAOA, QSVM/QNN) with classical baselines; mentored hybrid classical–quantum workflows and parameter sweeps; enforced reproducibility (clean repos, env pinning, MLflow) and fair-comparison protocols).
  • DSE 3010 (ML for DS) – Fall 2024: Taught model selection and diagnostics; standardized repos/environments and MLflow tracking with diagnostic checks (ablations, error analysis, leakage/imbalance).
• Feb 2022 – Sep 2022 Program Manager (Data Strategy Development, Predictive Analytics & Business Intelligence) — ShopUp, Dhaka, Bangladesh
  • Built B2C/B2B analytics; reduced return ratio via data-driven operations; developed forecasting and BI dashboards.
• Jun 2020 – Feb 2022 Project Manager (Data Analytics, SQL Database Management, and Agile Leadership) — Icon Information Systems Ltd., Dhaka
• Nov 2018 – May 2020 Assistant Project Manager — Millennium Information Solution Ltd., Dhaka
• Mar 2017 – Jul 2018 Assistant Project Coordinator — ESOL BD Limited, Bangladesh

Education

• Oct 2022 – Mar 2024 M.S., Information Technology — Washington University of Science and Technology (WUST), Alexandria, VA
• 2013 – 2018 B.S., Computer Science and Engineering — North South University, Dhaka

Online Certificates

• Microsoft Power BI Data Analyst Specialization — Microsoft
• Google Advanced Data Analytics — Google
• Google Business Intelligence — Google
• Google Data Analytics — Google
• SAS Statistical Business Analyst — SAS
• Salesforce Sales Operations — Salesforce
• Project Management & Other Tools for Career Development — University of California, Irvine

Skills

Programming & Platforms: Python (PyTorch, scikit-learn, NumPy, Pandas) · Git/Linux · Jupyter · Docker (basic) · GPUs · cloud · edge/IoT

Trustworthy ML & Stats: Transformers · GNNs · multimodal fusion · robust training · calibration & uncertainty · ablations & error analysis

Genomics & Functional Genomics: Multi-omics integration · CRISPR/perturbation data · ATAC-seq & regulatory signals · gene & target prioritization

Quantum & Optimization: Qiskit · VQE/QAOA · hybrid classical–quantum workflows · QUBO/Ising mappings · quantum-guided hyperparameter optimization

Privacy & Cryptography: Homomorphic Encryption (CKKS, BGV; OpenFHE/SEAL) · differential privacy · secure aggregation · post-quantum cryptography (KEMs & signatures; liboqs) · threat modeling

Data Engineering & Reproducibility: SQL · ETL pipelines · schema design · MLflow/W&B · environment pinning · unit tests · CI/CD (GitHub Actions) · DVC

Visualization & Writing: Matplotlib · dashboards · LaTeX/Overleaf · figure & table design

Domains: Regulatory genomics · functional genomics (CRISPR & epigenomics) · multi-omics integration · clinical prediction