Thanks to visit codestin.com
Credit goes to atharvasathaye.github.io

MS Information Management · UIUC · 2026

I turn messy data into
measured answers.

Data Analyst with 2.5 years at Accenture across automotive operations and healthcare insurance, plus independent analytics work. I work in SQL, Python, and Power BI, turning ambiguous business questions into answers through funnel, retention, and experiment analysis.

2.5 yrs industry experience SQL, Python, Power BI primary stack Open to opportunities

Work

Selected Projects

End-to-end builds — from raw data ingestion to deployed dashboards and APIs.

GitHub ↗

E-Commerce Growth Analytics

Modeled GA4 event data into a dbt star schema in BigQuery under 37 tests. Trained a gradient boosting churn model at 0.784 AUC with SHAP attribution. Sized $83K monthly revenue at risk through cohort retention and RFM segmentation. Designed a powered A/B test and measured a layout change worth $24K per month using difference-in-differences.

SQL BigQuery dbt Python Looker Studio A/B testing
GitHub ↗

Digital Banking Competitive Benchmark

Analyzed 598,000+ CFPB complaints and 2M+ app store ratings across 6 US banks, normalized by deposit base. Found the three largest banks carry 172K–183K complaints each against 15K–32K for peers. Scored competitive gaps on RICE and surfaced priority improvement areas.

Python CFPB API NLP Git
GitHub ↗

Healthcare Billing Chatbot

AI billing assistant with RAG retrieval over insurance policy documents, built for a real client and presented to the CEO and executive team. Reduced hallucination rates 40% against a held-out question set. Bilingual (EN/ES) with PHI/PII guardrails and Azure deployment.

FastAPI OpenAI RAG Azure LangChain
GitHub ↗

Supply Chain Inflation Forecasting

Live pipeline pulling FRED CPI sub-indices and NY Fed GSCPI. Granger causality tests across 12 lags with significant results at p < 0.0001. VAR(p) model forecasts inflation 12 months ahead. Deployed as a public Streamlit dashboard with monthly GitHub Actions refresh.

Python Streamlit statsmodels FRED API GitHub Actions
GitHub ↗

Career Insights Dashboard

Star schema data warehouse tracking UIUC graduate employment outcomes. SQL ETL pipeline feeding Tableau dashboards. Recognized by faculty and recommended to the university career center.

SQL Tableau Star Schema ETL
GitHub ↗

VeriStream AI

Video verification API for deepfake detection built for UIUC's COZAD New Venture Challenge. Runs HuggingFace frame-level classifiers, Laplacian blur gating, face detection, and ffprobe metadata inspection to produce a JSON risk report with trust score and frame timeline.

Python FastAPI PyTorch HuggingFace OpenCV

Toolkit

Skills & Stack

Languages

Python R SQL JavaScript HTML / CSS Bash

ML & Data

scikit-learn PyTorch HuggingFace pandas NumPy statsmodels

Visualization

Tableau Power BI Plotly Altair Seaborn Streamlit

Backend & Cloud

FastAPI PostgreSQL Azure Git Docker GitHub Actions

Background

Education & Experience

Graduate-level coursework in data engineering, HCI, and statistical modeling, alongside applied project work.

2024 – 2025

MS Information Management — UIUC

Coursework in data warehousing, HCI, statistical analysis, information modeling, and data visualization. GPA focus on applied project delivery.

2024

Applied Business Research — Healthcare Client

Team of 5 building an AI-powered billing assistant and mock patient portal. Delivered bilingual chatbot with RAG retrieval, PHI guardrails, and Azure deployment.

2023 – 2024

COZAD New Venture Challenge — UIUC

Built VeriStream AI, a deepfake detection API, as part of UIUC's premier startup competition. Designed full architecture, ML pipeline, and Streamlit dashboard.

2023

Haptics Research — UIUC

Research assistant on cross-modal music experience for Deaf and Hard of Hearing users. Contributed to music transcription experiments and wearable haptics prototyping.

Let's build something together.

Open to full-time roles in data science, ML engineering, and analytics. Based in Champaign, IL — open to relocation.