Dev Sanjay Shah
Turning raw data into clear, decision-ready insight for product, operations, and strategy teams.
About
I’m Dev Sanjay Shah, a recent M.S. Information Systems graduate from SUNY Binghamton.
I turn messy operational data into clean, decision-ready insights — whether that means building cohort retention models, designing SQL warehouses, or prototyping dashboards that help stakeholders ask better questions.
My sweet spot is at the intersection of business context and technical execution: writing queries that matter, validating assumptions with statistics, and communicating results in a way that moves projects forward.
Outside of data, you’ll probably find me hiking a new trail, planning my next trip, or working through a calisthenics routine — I’ve been at it for six years and counting. I’m big on fitness, curious about new places, and the kind of person who believes good analysis and a good laugh can coexist.
Case studies
End-to-end analytics work: problem definition, modeling, visualization, and a clear story. Each repo is open for review.
A full analytics warehouse for SaaS revenue and retention. Built a star schema in DuckDB, modeled cohort retention, MRR waterfall, NRR, and LTV, and served metrics through a FastAPI + React dashboard.
End-to-end ML system for hospital readmission risk. PySpark medallion architecture, scikit-learn modeling with SHAP explainability, FastAPI inference, and a React frontend with RAG over PubMed via Qdrant.
Multi-page Streamlit dashboard benchmarking real CMS hospital surgical outcomes. Includes a calibrated synthetic patient-level layer for risk adjustment and predictive modeling.
Streamlit dashboard surveilling real CDC/CMS hospital-acquired infection data. Features funnel-plot control charts, FDR-corrected hypothesis testing, and a predictive model.
Enterprise research agent with hybrid RAG and validated NL-to-SQL over a local Llama 3.1 8B model. Every claim is cited to a document chunk or SQL result.
A churn analytics workbook and dashboard pipeline for exploring drivers of customer attrition, segmenting at-risk users, and quantifying retention lift.
Tech stack
A practical mix of querying languages, analytics libraries, and BI platforms I use across projects.
What I do
Every project I take on tends to land in one of these lanes — and usually crosses two of them.
Root-cause analysis on revenue, churn, and operational metrics using SQL, Python, and controlled comparisons.
Star-schema design, dbt transformations, and dimensional modeling that keeps reporting fast and trustworthy.
End-to-end ML workflows from feature engineering to model evaluation, with explainability baked in.
Interactive dashboards and executive summaries that make complex findings easy to act on.
Workflow
A repeatable path from business question to actionable insight — with the tools I use at each stage.
Start with the business question. I clarify the stakeholder need, success metric, and scope before touching any data.
Pull the right data from warehouses, APIs, or flat files using SQL and version-controlled transformations.
Clean, normalize, and reshape the data so downstream analysis is reliable and reproducible.
Apply descriptive and diagnostic techniques — cohorts, segmentation, hypothesis tests — to find the real story.
When prediction adds value, I build interpretable models and validate them before any production use.
Turn findings into dashboards and reports that stakeholders can read, filter, and act on.
Close the loop with clear, prioritized recommendations tied back to the original business question.
Analysts are hired to drive decisions, not just run queries. This process keeps every project tied to a business outcome, so the final deliverable is something stakeholders can act on.
Journey
2024 – 2026
State University of New York at Binghamton
Concentrated in data analytics, business intelligence, and systems design. Completed coursework in database systems, machine learning, and data-driven decision making.
Ongoing
Self-directed
Building end-to-end analytics projects spanning SQL warehouses, predictive modeling, healthcare outcomes, and interactive dashboards.
Next
Open to opportunities
Looking for a team where I can translate complex data into clear business impact.
Contact
Hiring for an analyst role, or just want a second pair of eyes on a dataset? Send a note — I read everything.
