Business Analyst Power Bi

Ganesh Pundalik Naik

New York, US +1-607-***-**** ******@******.*** LinkedIn

Professional Summary

Strategic and hands-on Data & Business Analyst with proven success in building impactful analytics solutions across finance and healthcare. Expertise in developing real-time fraud detection systems, predictive credit risk models, and regulatory dashboards using Python, SQL, Power BI, PySpark, and ML frameworks. Adept at transforming complex datasets into actionable insights and communicating them to business stakeholders through visual storytelling. Passionate about solving high-impact business problems with data, and known for driving measurable results through analytics, automation, and end-to-end project ownership. Technical Skills

Programming & Scripting Languages: Python SQL DAX PySpark Machine Learning & AI: Logistic Regression XGBoost Random Forest Decision Trees Gaussian Mixture Model K-Means PCA Isolation Forest CNN

(VGG16, ResNet50) LSTM AutoEncoder SHAP BLEU Score METEOR Feature Importance Hypothesis Testing A/B Testing t-tests Confidence Intervals Natural Language Processing & Computer Vision: Tokenization Word Embeddings Sequence Modeling Image Feature Extraction Transfer Learning Data Processing & ETL: PySpark (RDDs, DataFrame API) Pandas NumPy Data Cleaning Feature Engineering Encoding Scaling Data Visualization & BI Tools: Power BI (DAX, Dashboards) Tableau Matplotlib Seaborn Streamlit Excel (Pivot Tables, Charts) Big Data & Cloud Tools: Apache Spark AWS (S3, SageMaker - basic) Google Colab Jupyter Notebook Tools & Development Platforms: Git GitHub VS Code Anaconda Google Sheets APIs Business & Domain Relevance: Finance Analytics Healthcare Analytics Regulatory Compliance (Water System Project) Stakeholder Reporting KPI Dashboarding Documentation & Requirement Analysis Professional Experience

Data Analyst

Capital One Financial USA June 2024 – Present

● Developed a real-time fraud detection system using Python, achieving over 92% recall and <3% false positives by applying Isolation Forest and AutoEncoder models on simulated transaction data.

● Built a credit risk scoring model using XGBoost and LendingClub financial data, achieving 0.91 ROC-AUC, and leveraged SHAP values to ensure model transparency and explainability.

● Designed and executed an A/B testing simulation for product optimization (credit card application flow), resulting in a 5.3% increase in conversion rate through statistical testing (t-tests, confidence intervals).

● Cleaned and analyzed over 50,000+ records using SQL, Pandas, and NumPy; engineered features like debt-to-income ratio, credit history bins, and loan purpose categories.

● Created interactive dashboards and reports using Tableau, Matplotlib, and Streamlit to communicate insights from fraud models, A/B tests, and risk scoring to non-technical stakeholders.

● Applied end-to-end analytics workflows across fraud prevention, credit decisioning, and marketing experimentation — reflecting real business use cases at fintech firms like Capital One.

● Collaborated with business stakeholders during internship to build SQL-based dashboards in Tableau, improving reporting speed and stakeholder visibility across KPIs.

● Automated weekly reporting tasks using Python scripts, saving 8–10 hours per week and reducing manual errors in internal reporting workflows. Business Analyst Finance & Healthcare

Winfinite Info Solutions Pvt Ltd Pune, India Jan 2022 – July 2023

● Delivered end-to-end analytics projects modeled after real-world client scenarios across the banking and healthcare industries, focusing on fraud detection, credit risk, hospital readmission, and insurance fraud.

● Collaborated with a team of data professionals to define business objectives, gather functional requirements, and translate them into analytical workflows using SQL, Python, and BI tools.

● Built a credit risk prediction model using XGBoost and SHAP, achieving 0.91 ROC-AUC and producing explainable risk scores for loan approval decisions.

● Developed a real-time fraud detection system using anomaly detection techniques (AutoEncoder, Isolation Forest), achieving 92% recall and delivering real-time alerts via a Streamlit dashboard.

● Predicted 30-day hospital readmissions using Logistic Regression, identifying high-risk patients and enabling hospitals to implement early intervention strategies.

● Flagged anomalies in health insurance claims using pattern-based and statistical outlier detection; built Power BI dashboards to help fraud investigation teams prioritize review.

● Automated Excel-based reports using Python, reducing manual effort by 10+ hours per week and increasing reporting accuracy and turnaround time.

● Presented final dashboards and business recommendations to mock stakeholders in simulated client review sessions, demonstrating communication and data storytelling skills.

Educations

Master of Science in Information Systems Aug 2023 - May 2025 Binghamton University, State University of New York NY, USA Bachelor of Technology in Electronics & Telecommunications Engineering Aug 2019 - May 2023 Savitribai Phule Pune University (formerly University of Pune) Pune, India Projects

Image Captioning with Deep Learning Python, TensorFlow, Keras, VGG16, LSTM, NLP, BLEU Score

● Developed a deep learning model combining CNN (VGG16) and LSTM to generate human-like captions for images using the Flickr8k dataset.

● Achieved a 90% BLEU score, demonstrating high-quality caption generation for use cases in accessibility and automated media tagging.

● Implemented custom tokenization, word embeddings, and sequence modeling to maintain contextual and grammatical accuracy.

● Evaluated performance using BLEU/METEOR metrics through an end-to-end TensorFlow/Keras pipeline. Public Water Systems Segmentation for Regulatory Compliance PySpark, Python, Power BI, DAX, GMM, Logistic Regression, SQL

● Built a machine learning solution using Gaussian Mixture Models and Logistic Regression to classify public water systems into compliance risk clusters with 88% accuracy.

● Processed and transformed large-scale regulatory datasets using PySpark, improving pipeline scalability and performance.

● Developed an interactive Power BI dashboard using advanced DAX, enabling real-time visualization and risk filtering.

● Collaborated with stakeholders to define risk metrics, reducing analysis time by 50% and improving high-risk system identification by 90%. Certifications

● Google Data Analytics Professional Certificate

● Microsoft Power BI Data Analyst Associate (PL-300)

● IBM Data Analyst Professional Certificate

Contact this candidate