Turning DataInto DecisionsThat Move Metrics
I build analytics tools, ML models, and AI systems that turn business questions into answers worth acting on. I work where analytics, machine learning, and business strategy overlap, building things that hold up in front of executives and stay useful to the people running the business.
Introduction
Overview
I'm a Master's student in Business Analytics at Purdue with about a year and a half of industry experience. I've built dashboards that ran on the executive floor, models that recovered margin nobody knew was leaking, and most recently, multi-agent AI systems that automate the kind of reasoning a finance committee usually does in a room.
Business Analyst
Turning messy data into decisions stakeholders can defend.
Data Analyst
SQL, Python, and dashboards that make the pattern obvious.
Analytics Consultant
Framing the right question before reaching for the model.
Applied AI Engineer
LangGraph multi-agent systems and ML pipelines that ship.
Selected Work
Projects.
Problem → Approach → Insights → Impact. Each card opens a full case study with the methodology, code, and quantified outcome.
Applied AI / Corporate Finance · 2025
Multi-Agent Financial Decision System
AI agents that simulate a finance committee
~30%
of reasoning errors caught by adversarial agent review
Built a LangGraph system where four agents (Strategy, Risk, Finance, Modeling) argue with each other to produce scenario-based corporate finance recommendations.
Financial Risk / ML · 2024
Bankruptcy Prediction Ensemble
Stacked gradient-boosting model for financial distress
0.9056
AUC on 8K companies × 64 financial ratios
Two-level stacking ensemble (LightGBM, XGBoost, CatBoost feeding into Logistic Regression) reaching 0.9056 AUC on 8,000 companies and 64 financial ratios.
Marketplace Analytics · 2025
Airbnb Host Retention Optimization
Allocating $1,000 retention gifts to maximize profit, not accuracy
10.6%
modeled profit lift on a 1,312-host test set
Built a profit-aware decision rule that, on a held-out test set of 1,312 hosts, would have lifted projected profit by 10.6% by gifting only the 23.4% of hosts where retention dollars actually pay back.
Retail Analytics · 2025
Customer Segmentation & Churn Prediction
Finding the 28% of customers who drive 83% of revenue
82.8%
of revenue concentrated in top 28.9% of customers
Segmented 5,878 customers from 779K transactions with RFM features, then layered a churn classifier on top so the marketing team knew who to chase first.
My Journey
Work Experience
Student Intern (Industry Practicum)
Microsoft
- Designed a Community Health Score with 4 dimensions (engagement, quality, responsiveness, risk) tracking 60+ open-source projects across Reddit, GitHub, and Stack Exchange.
- Used sentiment analysis and topic modeling on roughly 80K posts and issues to surface what was actually moving each community up or down.
- Shipped a Power BI dashboard that flagged 12 at-risk communities for the program team to look into first, saving them several days of manual triage per quarter.
Jan 2026 – Apr 2026
West Lafayette, IN
Jan 2024 – Jul 2025
Indore, India
Junior Analyst
3E WebApps Pvt. Ltd.
- Owned 6 Power BI dashboards covering sales, operations, and customer KPIs that 20+ stakeholders relied on weekly.
- Automated recurring performance reports and cut manual reporting time by about 30%, freeing the team for actual analysis.
- Worked with marketing and supply-chain leads on stock planning, cost control, and loyalty campaigns. The cohort retention work became the basis for the loyalty program rollout.
- Reorganized KPI definitions across teams so the same metric meant the same thing in every room. Cut the volume of 'why does this number not match' threads in half.
Business Analyst & Market Research Intern
Select Brands Pvt. Ltd.
- Tuned Amazon ad spend across 40+ SKUs and brought ACOS down by about 20% over 3 months.
- Ran competitor and category research that supported the launch of 10 new self-care products.
- Found and sized 3 product gaps in the category that the team built into the next quarter's roadmap.
Jun 2023 – Oct 2023
Indore, India
Education
Master of Science in Business Analytics & Information Management
Purdue University — Daniels School of Business · West Lafayette, IN
December 2026
Bachelor of Business Administration in Information Technology
Symbiosis International University · Pune, India
June 2023
Toolbox
Skills.
A working stack — analytics tooling, ML libraries, and the cloud environments where it all runs in production.
Analytics & BI
Data & Databases
Programming
ML & AI
Cloud
Methods
Certifications & publications
- ·AWS Certified Cloud Practitioner — Amazon Web Services
- ·Microsoft Azure AI Data Fundamentals — Microsoft
- ·Research paper — Bias-Variance trade-off in ML algorithms — Peer-reviewed journal
Tools & Tech
Day-to-day stack
- Python
- SQL
- Power BI
- Tableau
- AWS
- Google Cloud
- MongoDB
- Databricks
- scikit-learn
- LangChain
I would love to hear from you