Data Analyst who turns messy data into decisions that stick.

$48Bn+ in US Medicare drug spend with no structured view of which categories, manufacturers, or brands were driving it.

• Accomplished end-to-end pharma market analysis by querying 100,000+ FDA Medicare records across 1,200+ brands using Python and 6 advanced SQL queries (CTEs, LAG, window functions), which resulted in identifying oncology and immunology as the top spend categories within $48Bn+ in annual US drug expenditure.
• Accomplished manufacturer concentration risk mapping by building a 6-chart Power BI dashboard and 3-slide consulting deck, which resulted in a simulation-format deliverable replicating real pharma advisory output.

CBRE's leasing teams had no unified view of office demand, vacancy compression, and corridor risk across Gurugram NCR micro-markets.

• Accomplished end-to-end real estate market intelligence by integrating 6 data sheets covering office leasing, residential trends, vacancy rates, and early warning signals into a unified Tableau dashboard, which resulted in a single source of truth for CBRE's Gurugram NCR advisory workflow.
• Accomplished corridor-level demand-supply gap analysis by engineering a calculated gap field across key micro-markets, which resulted in identifying Cyber City as the most supply-constrained corridor with a -200K sqft gap signalling acute leasing pressure.

Lenders had no structured early warning system for LTV breaches or margin calls when collateral value drops under market stress.

• Accomplished credit risk tier segmentation across 300,000+ client records by engineering 4 risk tiers using Python and SQL, which resulted in a structured credit line review framework aligned with institutional lending standards.
• Accomplished collateral stress-testing by simulating 10–30% market declines across equities, bonds, and mutual funds in Excel and Power BI, which resulted in automated LTV breach detection and margin call threshold alerts.

Pharma stakeholders lacked phase-wise visibility into trial success rates and therapeutic area performance across a 400,000+ record pipeline.

• Accomplished phase-wise drug pipeline visualisation by processing 400,000+ clinical trial records using Python, Tableau, and Seaborn, which resulted in surfacing drop-off rates at each trial phase and identifying therapeutic areas with the highest trial-to-approval conversion.

India's trade and debt data across strategic partners was spread across 9+ fragmented sovereign sources with no unified analytical view.

• Accomplished sovereign debt and trade flow consolidation by integrating 100,000+ records from 9+ government and international sources into a unified Power BI data model, which resulted in a strategic intelligence dashboard mapping India's key bilateral partner dependencies and concentration risks.

Students and counsellors had no data-driven tool to predict seat allocation outcomes across medical colleges based on rank and category.

• Accomplished predictive seat allocation modelling by building a Python and Power BI solution on 50,000+ student records, which resulted in 92% prediction accuracy for counselling outcomes across medical colleges.

A global restaurant chain needed performance benchmarking across 9,551 outlets in 15+ countries to identify regional underperformers.

• Accomplished multi-country restaurant performance analysis by processing 9,551 outlet records across 15+ countries using Power BI, which resulted in regional benchmarking insights and underperformer identification aligned with Accenture's analytics consulting framework.