Consumer Affairs Lead Conversion Case Study
Built a predictive modeling pipeline (Logistic Regression, XGBoost), achieving ROC-AUC ≈ 0.68 and 2.3–2.4× lift in top-decile leads; shipped Power BI dashboards for stakeholders.
I build end-to-end data products—from ingestion and modeling to deployment and dashboards—focused on reliability, scalability, and measurable outcomes that drive smarter decisions.
I design reliable, measurable data solutions—from ingestion and feature engineering to ML deployment and decision dashboards. I turn ambiguous business questions into outcomes using Python, SQL, Databricks, Snowflake, and modern cloud/MLOps stacks.
Data wrangling, automation, and statistical programming
KPI dashboards, business storytelling, and data-driven insights
Predictive modeling, A/B testing, and generative AI solutions
Scalable data pipelines, model deployment, and orchestration
Data integration, warehousing, and cloud-native analytics
Version control, agile teamwork, and rapid prototyping
Built a predictive modeling pipeline (Logistic Regression, XGBoost), achieving ROC-AUC ≈ 0.68 and 2.3–2.4× lift in top-decile leads; shipped Power BI dashboards for stakeholders.
Designed AI agents for search & optimization (maze, n-Queens, Connect-4) using BFS/DFS/A*, Hill-Climbing, Simulated Annealing, and Minimax+Alpha-Beta. Benchmarked heuristics and decision efficiency under adversarial settings.
Compared supervised learners (LogReg, PCA+MLP, CNN, Wide & Deep). Tuned optimizers (RMSProp, AdaDelta) and evaluated across structured, image, and text datasets.
Designed A/B tests (Google vs Facebook), analyzed KPIs, and improved marketing ROI by 22% with budget allocation recommendations.
Forecasted Shopify & Alibaba using ARIMA & exponential smoothing to support investor risk profiling.
Performed data wrangling, EDA, and visualization on Super Bowl commercials to uncover trends in ad spending, brand presence, celebrity appearances, and viewer engagement using Python and visualization libraries.
This repository captures my Python learning path and a final project analyzing job-market data. It includes hands-on notebooks for core Python (data types, control flow, functions), plus pandas/NumPy data wrangling and Matplotlib/Seaborn visualization. The capstone cleans raw postings, engineers skills/keyword features, explores demand by role/tech and salary ranges, and presents insights through clear, reproducible notebooks.
Predicted hourly demand (92%+ accuracy) with weather & trip features; trained LightGBM, XGBoost, and CatBoost.
Reduced holding cost by 72% and optimized $2.7M unsold stock via Power BI + Python analytics workflows.
Cut research time by 40% with a LangChain/FAISS assistant for semantic search across financial news & docs.
CVR College of Engineering — Hyderabad
Southern Methodist University — Dallas
I'm actively seeking opportunities in Data Science and AI. Let's discuss how I can contribute to your team!