Pavani Satuluri Data Analyst-| - | USA | LinkedIn
Summary
Experienced Data Analyst with 2+ years of strong background in data manipulation, analysis, and visualization across multiple industries.
Skilled in SQL, Python, R, Power BI, and Tableau, with expertise in using advanced statistical techniques and business intelligence tools to
extract insights from large datasets. Adept at creating interactive dashboards, improving data accuracy, and driving data-driven decisionmaking. Proven ability to optimize processes, enhance operational efficiency, and deliver actionable insights to senior stakeholders,
contributing to business growth and cost reduction.
Technical Skills
•
•
•
•
•
•
•
Programming Languages: Python (Pandas, NumPy, Scikit-learn, Statsmodels), R, SQL
Tools & Technologies: Power BI, Tableau, AWS (S3, Lambda), Azure (SQL Database, Data Factory), DAX, SciPy
Data Analysis: Time-Series Analysis, Regression Analysis, PCA, Survival Analysis, Fraud Detection Models, Hypothesis Testing
Cloud Platforms: AWS, Azure
Data Validation & ETL Processes: CTEs, Window Functions, Data Cleansing, LOD Expressions
Machine Learning: Decision Trees, Logistic Regression
Data Storage & Processing: Amazon S3, Azure Data Lake
Professional Experience
Data Analyst, Frost Bank
•
•
•
•
•
•
•
01/2024 – Present | TX, USA
Worked on Fraud Detection and Prevention Analysis, applying agile methodologies and collaborating with cross-functional teams to
identify and mitigate fraudulent transactions, resulting in a 15% reduction in fraud-related losses over six months.
Utilized SQL methods like window functions (ROW_NUMBER, RANK) and Common Table Expressions (CTEs) to analyze large-scale
transaction data from Frost Bank’s internal systems, identifying anomalous patterns and enhancing fraud detection accuracy by 20%.
Implemented advanced data analysis techniques, including time-series analysis and statistical hypothesis testing, to identify irregular
transaction patterns, providing actionable insights to prevent fraudulent activities and reducing false positives by 10%.
Applied Python (Pandas, NumPy, Scikit-learn, and Statsmodels) for data manipulation, feature extraction, and implementing fraud
detection models, including decision trees and logistic regression, resulting in a 25% improvement in fraud detection rates.
Ensured data accuracy through robust validation using DAX and custom logic, cross-referencing transaction data for inconsistencies.
Validated over 500,000 transaction records to ensure integrity and precision, reducing data errors by 8%.
Created advanced Power BI dashboards with features like drill-through, dynamic slicers, and real-time updates, presenting fraud
trends and KPIs to senior stakeholders, which led to a 12% improvement in proactive fraud detection.
Utilized AWS tools like Amazon S3 for scalable data storage and AWS Lambda for real-time data processing, streamlining data
ingestion and fraud detection workflows, enabling faster decision-making and reducing processing time by 30%.
Associate Data Analyst, CitiusTech
•
•
•
•
•
•
•
08/2022 – 12/2023 | TX, USA
Conducted broad cost analysis for 300+ healthcare clients, identifying inefficiencies in treatment, medication, and hospitalization
workflows. Delivered insights led to a 15% drop in healthcare costs, directly impacting over $50 million in annual spending.
Extracted and transformed healthcare data from Azure-based data lakes using Azure SQL Database and Azure Data Factory for ETL
processes. Employed advanced SQL methods such as Common Table Expressions (CTEs) and Window Functions to optimize complex
queries, reduce data processing time by 40% and query execution from 20 minutes to 12 minutes.
Organized multi-variable regression analysis to model cost drivers for over 50k patient records. Applied Time Series Decomposition to
forecast future treatment costs, improving prediction accuracy by 18%, enabling more precise budgeting for healthcare providers.
Automated data manipulation processes using Python, Pandas for data cleansing, NumPy for large-scale numerical computation, and
SciPy for statistical analysis. This improved the speed of cost report generation by 45%, enabling quicker decision-making.
Employed R for performing PCA on a $10 million cost dataset to reduce dimensionality and reveal hidden trends. Additionally,
Survival Analysis used to predict high-risk patient groups, improving resource allocation for 5k high-risk patients.
Designed and implemented rigorous data validation checks using LOD expressions in Tableau, ensuring data consistency across
diverse hospital departments. These validation steps eliminated data discrepancies, increasing reporting accuracy by 98%.
Developed interactive dashboards in Tableau integrating real-time patient data, geospatial analysis, and predictive trend lines. These
dashboards, serving over 100 senior executives, improved strategic decision-making speed by 25%, providing actionable insights that
directly impacted clinical and financial outcomes.
Education
Master of Science in Information Technology and management
The University of Texas at Dallas
08/2022 – 12/2024
TX, USA
