Home About Me Portfolio Resources Blog Connect on LinkedIn →

Data Scientist & ML Engineer

Turning data
into decisions
that matter

I build machine learning models, interactive dashboards, and NLP systems that give teams the clarity to act — not just analyze.

Featured Portfolio All Projects
Suparna Chowdhury PDF · Updated 2026
View Resume
Tech Stack
Python SQL Power BI Tableau XGBoost Scikit-learn NLP Feature Engineering MLOps GAN RAG Systems Reinforcement Learning
Featured Project
Data Governance & Compliance Risk Dashboard
Translated 5,500 audit records into a financial exposure framework — quantifying compliance gaps in euros for executive-level decision-making.
Power BI DAX Risk Modeling
Profile

A data scientist who speaks business

I translate data into decisions that drive real business results — bridging technical innovation and strategic thinking across machine learning, predictive modeling, and NLP.

From building robust models in Python to designing A/B tests and crafting dashboards in Tableau and Power BI, I deliver cost-effective solutions grounded in customer needs. My peer-reviewed publications reflect a commitment to staying at the forefront of the field.

My toolkit spans Python, SQL, Power BI, Tableau, and Excel — but what sets me apart is translating technical complexity into narratives that resonate with both engineers and executives.

Off the clock, I play piano — and whether it's data or music, I'm always finding patterns in the noise.

Education
MS in Electrical Engineering
University of South Alabama · July 2015
Certifications
IBM Data Science Professional Certificate
IBM · January 2023
Verify credential ↗
Fundamentals of Visualization with Tableau
UC Davis · January 2023
Verify credential ↗
Find me online
GitHub Blog LinkedIn Twitter
Research

Published Work

Peer-reviewed research spanning reinforcement learning, computer vision, and energy systems.

IEEE · December 2017
A reinforcement learning algorithm based technique for thermal energy management of a PEM fuel cell power plant
Read paper
SPIE · April 2016
Efficient face recognition using local derivative pattern and shifted phase-encoded fringe-adjusted joint transform correlation
Read paper
Services

What I can
do for you

From raw data to boardroom-ready insights — I cover the full journey, technically and strategically.

  • SQL queries and automated data extraction pipelines
  • Advanced transformation and cleaning with Python / dbt
  • Statistical hypothesis testing and KPI analysis
  • Trend identification and predictive forecasting
  • Interactive Tableau and Power BI dashboards
  • Executive-level data storytelling for stakeholders
  • DAX calculations and advanced data modeling
  • Custom visual analytics solutions
  • Supervised and unsupervised learning models
  • Predictive modeling for churn, LTV, and credit risk
  • NLP pipelines for sentiment analysis and classification
  • Model deployment and MLOps integration
  • Custom prompt engineering for LLM optimization
  • Building RAG (Retrieval-Augmented Generation) systems
  • Developing AI agents for task automation
  • Exploring AGI-adjacent frameworks and neural architectures
Work

Featured Portfolio

A curated look at high-impact work — from credit risk modeling to urban pattern discovery.

Data Governance Dashboard
Power BI · DAX · Risk Modeling
Data Governance & Compliance Risk Dashboard
Translated 5,500 audit records into a financial exposure framework, quantifying compliance gaps in euros for executive action.
View Project
Credit Risk
Python · XGBoost · Power BI
ML-Powered Credit Risk Assessment
93% accuracy on 32K+ loans. Identified $77M in default risk via XGBoost and a live Power BI dashboard.
View Project
Clustering
Python · K-Means · PCA
Urban Pattern Analysis via K-Means Clustering
Segmented 300 cities into archetypes using 8 socioeconomic indicators. Silhouette score: 0.374.
View Project
SQL Retail
SQL · Cohort Analysis · Retail
Retail Sales, Retention & Cohort Analysis
Uncovered sales trends, retention patterns, and discounting impact using advanced SQL on a real retail dataset.
View Project
Tableau DZV
Tableau · Dashboard Design
Mastering Dynamic Zone Visibility in Tableau
Two-part article series on building clean, interactive Tableau dashboards with DZV, Parameter Actions, and Sets.
Read Article
View All Projects
Expertise

Areas of Interest

The disciplines I work in — and genuinely care about.

Machine Learning
Machine learning is a tool, not a buzzword. I love optimizing algorithms and solving real business problems.
Natural Language Processing
With NLP, I extract insights from messy text and turn human language into structured, actionable knowledge.
Data Visualization
Data speaks louder when visualized. I design charts and dashboards that drive decisions, not just look good.
Statistical Analysis
I use statistical techniques to validate assumptions and guide decision-making with confidence and rigor.
SQL Database
SQL is my tool for organizing data, running efficient queries, and making sense of massive, complex datasets.
Data Preprocessing
I treat data preprocessing as an art — refining raw data into clean, efficient structures ready for modeling.