NARAYANAN KRISHNAMURTHY
Sunrise, FL ***** 412-***-**** ******@*****.*** LinkedIn Profile GitHub Portfolio
SUMMARY
Senior Data Scientist leveraging advanced analytics, predictive modeling, and SQL/Python expertise to drive business insights in regulated, data-rich environments. Delivered measurable improvement in self-service analytics adoption, precision document intelligence, and revenue protection strategies. Skilled in collaborating cross-functionally to translate complex data into actionable results.
CORE SKILLS
• Data Science & Modeling: Predictive Modeling, Statistical Modeling, Machine Learning (ML), Natural Language Processing (NLP), Time-Series Analysis, Clustering, Random Forest, SVM, Multivariate Regression, Anomaly Detection, Topic Modeling, Change Point Analysis, Hypothesis Testing, Model Validation, Model Tuning, Feature Engineering, Tree-based Methods, Performance Metrics, Data Science
•Programming & Tools: Python, SQL, R, Scikit-learn, PyTorch, Pandas, Teradata, Postgres SQL, Microsoft Azure, Databricks, Tableau, Power BI, MS SQL Server, C#/.NET, Docker, Git
•Data Engineering & Platforms: Data Wrangling, Data Visualization, FedRAMP / Restricted Compute Environments, Cloud-based Systems, Synthetic Data Generation (SDV), Record Linkage, Semantic Search, Text Analytics
WORK EXPERIENCE
Internal Revenue Service (IRS) Mar 2021 - Present
Statistician Plantation, FL
Built SQL, R, Python, and Tableau solutions that improve workload visibility, self-service analysis, and secure AI/ML experimentation across LB&I practice areas.
•Streamlined workload visibility across ~11M corporate and partnership returns by engineering a reproducible data wrangling pipeline in SQL, R, and Python with pandas, supporting a self-service Tableau dashboard that reduced ad-hoc requests and enhanced business decision-making.
•Applied statistical modeling and advanced trend analysis on SEC 10-K MD&A filings using NLP and text analytics to identify emerging issues, enabling business insight generation and supporting strategic project recommendations.
•Enabled privacy-safe AI/ML development in FedRAMP-restricted cloud-based systems by designing and validating synthetic corporate tax return data with Python SDV and applying A/B tests and calculationg KPI to ensure data fidelity, leading to model experimentation without exposing live taxpayer data.
•Expanded adoption of analytical tools across 3 communities of practice serving 50–150 analysts weekly by presenting solutions in internal and cross-agency forums, leading to faster onboarding and broader use of R, Python, SQL, and AI.
Frontier Communications Jul 2020 - Mar 2021
Data Science Analyst Coconut Creek, FL
Analyzed large-scale call center data using R and Teradata SQL to create reports identify call patterns, strengthen customer-to-call matching, and improve reporting reliability across a 10-month analysis window in 2020.
•Improved customer record linkage accuracy by ~2% through Teradata SQL and pandas workflows that connected call records to customer accounts using caller ID and related identifiers, leading to reliable customer-level analysis across large call logs.
•Refined customer-to-call matching by leveraging attributes such as caller ID, partial addresses, and multiple phone numbers within Teradata-based analysis, leading to stronger match quality and more dependable tracking of caller activity.
•Analyzed operational call patterns leveraging R and Teradata SQL to identify trends, frequent callers, and key drivers, informing business reporting and operational improvement.
•Automated daily forecast reporting by building a C#/.NET and Microsoft Task Scheduler workflow for Excel-based outputs, leading to more consistent report generation and reduced manual reporting effort.
•Performed clustering analysis using scikit-learn to segment caller groups and applied hypothesis testing to validate segment differences, enhancing targeted support strategies and improving performance metrics.
Fabmatics USA (Client: Magic Leap) Feb 2019 - Dec 2019
Software Engineer Plantation, FL
Supported Magic Leap’s AR glasses prototyping environment by analyzing gRPC, PLC, and Linux system logs, coordinating issue translation between client and offshore German engineers, and helping sustain day-to-day operations across two
800-meter multistage processing lines.
•Bridged communication and issue resolution between Magic Leap, offshore German engineers, and 2 on-site technicians by serving as the primary technical liaison, leading to clearer technical communication and smoother problem resolution.
•Sustained operational continuity across a multi-vendor production assembly platform by analyzing gRPC, PLC, and Linux system logs, leading to faster identification and support of daily manufacturing issues.
Cognizant Technology Solutions (Client: Florida Power & Light)
Jun 2018 - Feb 2019
Data Science Analyst Jupiter, FL
Supported FPL’s Miami revenue protection team and data leads by applying R, Python, SPSS, and Postgres SQL to smart meter data, helping strengthen anomaly analysis and workflow-ready analytical outputs across a high-volume utility environment.
•Enhanced revenue analytics and protection analysis across ~2 million customer accounts by applying predictive modeling and year-over-year time series analysis to smart meter data, improving detection of abnormal usage behavior impacting revenue.
•Improved workflow readiness by building repeatable analysis on Postgres smart meter data, leading to a more usable and structured process for anomaly evaluation.
•Embedded change point analysis into smart meter evaluation by identifying where usage patterns materially shifted over time, leading to stronger anomaly detection within the existing workflow.
ORAU (Client: DOE National Energy Technology Laboratory) Feb 2016 - Jun 2018
Data Science Analyst Pittsburgh, PA
Helped Principal Investigators and research collaborators by applying machine learning and statistical analysis to complex materials datasets, guiding alloy evaluation and research exploration.
•Enabled more structured evaluation of 9Cr steel test data by building R- and Python-based analytical tools across DOE and Japanese datasets, leading to stronger exploration of alloy properties and experimental results.
•Strengthened alloy-property analysis by applying multivariate regression, PCA, variable importance analysis, and random forest-based modeling to existing test data, leading to better assessment of composition-property relationships.
•Performed clustering analysis on alloy to identify specific segments to better model properties within segments
•Aligned DOE and Case Western collaborators through research discussions and a white paper, improving interpretation of complex analytical findings.
EDUCATION AND CERTIFICATIONS
University of Pittsburgh
Master of Science, Bioengineering
Pittsburgh, PA
George Washington University
Master of Science, Computer Engineering Washington, DC
•GPA: 3.7 GPA
•Achievements: Teaching Fellow for Circuit Design, Signals and Systems, Digital and Analog Communication, and Stochastic Processes
•Coursework: Electronics, Databases & JDBC
Regional Engineering College Warangal (now National Institute of Technology Warangal)
Bachelor of Technology, Mechanical Engineering
CERTIFICATIONS
• Applications of Gen AI, John Hopkins University: Certificate (https://www.mygreatlearning.com/certificate/PUDGNRYB), Transcript (https://drive.google.com/file/d/1foN07jopU4TvXeMXuEefABZVCT0rjv7Q/view) (JAN 2026)
•Making Data-Driven Decisions, MIT IDSS, Schwarzman
College: Certificate (https://www.mygreatlearning.com/certificate/KVYFRYEW), Transcript (- https://drive.google.com/file/d/1PmbdRrWsg67VPFLxojhweIvLX-qaN5Hq/view) (MAR 2025)
•Certificate in Computing: CDAC, India
HONORS & FELLOWSHIPS
•NIH Multimodal Imaging Fellow Award:2012–2013
•Teaching Fellow:Circuit Design, Signals and Systems, Digital and Analog Communication, and Stochastic Processes
Andhra Pradesh, India