Nagul
Shaik

Analyzed daily S&P 500 market data to uncover price trends, sector rotation patterns, and volatility clustering. Built automated reporting pipelines that pull live market data, compute moving averages (SMA 50/200), and surface anomalies — all visualized in an interactive Power BI dashboard with drill-down by sector, time period, and individual ticker.

Built a full churn prediction pipeline on Travelers Insurance policyholder data — from raw feature engineering to a production-ready Logistic Regression + Random Forest ensemble (81% accuracy, AUC 0.69). Identified top churn drivers (policy tenure, claim history, premium changes) and delivered cost–benefit analysis that informed a targeted retention campaign worth an estimated 21% revenue uplift.

Designed and deployed a comprehensive analytics solution on Centene-scale Medicaid/Medicare claims data. Used advanced SQL (CTEs, window functions, multi-table joins across eligibility, medical, pharmacy, and provider tables) to detect utilization anomalies, cost trend outliers, and care gaps. Power BI dashboards provided real-time visibility into member health outcomes and provider performance.

Built an end-to-end ML demand forecasting system simulating a Walmart-scale retail scenario (50+ product categories, 500 store locations, 3 years of sales history). Used a Prophet + XGBoost ensemble with holiday effects, promotions, and macroeconomic indicators as external regressors. Live Streamlit dashboard allows supply chain teams to run scenario simulations in real-time.

Designed a two-stage fraud detection pipeline on a 500K synthetic credit card transaction dataset. Stage 1 uses Isolation Forest for unsupervised anomaly pre-screening; Stage 2 uses a calibrated Random Forest classifier trained on flagged transactions. Power BI streaming dashboard surfaces high-risk alerts in near real-time, with drill-down by merchant category, geography, and customer profile.

Built an NLP pipeline that scrapes Reddit/Twitter brand mentions, applies VADER + transformer-based sentiment scoring, and correlates weekly sentiment shifts to product revenue movements. Discovered a consistent 3-week leading indicator: negative sentiment spikes predicted revenue decline with 82% directional accuracy — enabling marketing teams to respond proactively before impact hits the income statement.

Led business analysis for a regional bank's migration from a 15-year-old core banking platform to a cloud-native system. Produced comprehensive BRD (60-page), AS-IS/TO-BE process maps (22 workflows), stakeholder impact matrix, and phased implementation roadmap. Quantified ROI across 3 business units and secured C-suite sign-off through executive summary dashboard.

Performed a full pricing analysis for a B2B SaaS platform with 3,500 customers. Conducted cohort-based LTV analysis, feature adoption heatmaps, and price elasticity modeling across customer segments. Built a revenue simulation model in Python + Excel that allowed leadership to test 4 pricing tier structures and visualize MRR trajectory under each scenario before committing to a pricing change.

Developed a vendor risk intelligence model for a mid-market manufacturer recovering from COVID supply chain disruption. Built a weighted vendor risk scorecard (87 suppliers) across 6 dimensions: financial health, geographic concentration, lead time variance, quality compliance, dependency risk, and ESG footprint. Ran Monte Carlo simulations for 3 disruption scenarios to quantify revenue exposure.

Implemented a full RFM (Recency, Frequency, Monetary) segmentation model on a 200K+ customer e-commerce dataset. Combined RFM with a predictive CLV model (BG/NBD + Gamma-Gamma) to assign 12-month revenue forecasts to each customer. The resulting 5-tier segmentation strategy — each with tailored engagement recommendations — modeled a 34% improvement in marketing ROI by shifting budget from low-CLV to high-CLV acquisition channels.

Designed a comprehensive ESG intelligence framework for a Fortune 500 manufacturing firm preparing for SEC climate disclosure requirements (2024). Built a data aggregation pipeline pulling Scope 1/2/3 emissions, water usage, DEI metrics, governance scores, and regulatory filings into a unified Power BI dashboard. Gap analysis identified 8 compliance risks before the mandatory reporting deadline.

Conducted a full-scale Intelligent Process Automation (IPA) opportunity assessment for a mid-market insurance company. Used process mining techniques to analyze 12 high-volume business processes (claims intake, policy renewal, billing reconciliation), scored each on automation feasibility, built ROI models for 3 implementation waves, and created an AI adoption roadmap with vendor selection scorecard comparing UiPath, Automation Anywhere, and Microsoft Power Automate.

Full buy-side M&A financial model replicating bulge-bracket analyst work on a regulated utility acquisition. Covers DCF, trading comps (11 peers), precedent transactions (10 deals), merger accretion/dilution, sources & uses, and four sensitivity analysis tables — all cross-linked across 11 sheets.

Bulge-bracket quality sell-side M&A pitch book for GridPower's Board of Directors. Covers strategic rationale, financial highlights, football field valuation (6 methodologies), comparable company analysis, synergy waterfall, and full sale process timeline.