Naveen Sampath Email & Phone Number

Indianapolis, Indiana, Us

Indianapolis, Indiana, Us

• Analytical Solution: Deployed an interactive data-driven tool using ensembles of Kmeans and EM Clusters to identify market segments.• Data Evaluation: Wrote SQL and Python scripts to evaluate vendor data quality, integrating with in-house data assets that leveraged data models for better prediction and accuracy.• Experimental Design: Implemented bootstrapping on error analysis to identify the optimal minimum numbers of variables required for designing efficient experiment which reduced the research cost of around $2MM.• Communication: Explained analytical findings to improve awareness of machine learning techniques to higher management and anticipate cross functional collaboration.

Chicago, Illinois, Us

• System Failure Modeling: Collected CPU statistics from the server logs and predicted the failure of the servers with an accuracy of 71% two days in advance using spectrum-kernel Support Vector Machine(SVM).• Data Portfolio: Created and automated dashboard for server utilization and minimized system outages for critical research systems that saved a cost of 4MM/year on data recovery.• Cost Saving: Configured AWS S3 buckets with various life cycle policies to archive data to reduce cost up to 55% of AWS bills.

Chicago, Illinois, Us

Chicago, Illinois, Us

• Developed a cluster analysis model to optimize the box sizes and displayed POC to enable automation.• Analyzed the impact of standardizing box sizes used throughout the supply chain.

San Francisco, Ca, Us

• Generative Model: Built a generative adversarial network(GAN) to generate synthetic data from the limited dataset which increased the production model performance by 3%.• Anomaly Detection: Explored dimension reduction techniques like PCA, LDA and UMAP enhanced with DBSCAN and Isolation Forest to identify abnormal user behavior with an accuracy of 98%.• Data Processing: Helped designing a Data lake repository to maintain campaign data in the native format for advanced analytics.

Mumbai, Maharashtra, In

• Data Pipeline: Developed an ingestion pipeline using Apache Kafka to capture web events for Real-Time Market Analysis for airline dynamic pricing.• Deployment: Deployed the models on private cloud stack to enable the end users with SaaS capability.• Workflows: Designed and developed a workflow playbook for engineers to manage ETL jobs leveraging reusablecomponents.• Visualization: Created visualization dashboards for more than a dozen online reports helping clients identifyopportunities to improve profit.• Initiative: Built a POC on Data Mart focused on a specific business line that significantly improved the end-userresponse time without focusing on building complex queries.• Project Lead: Led the application development of two modules. Served as the primary contact for SEV-1 productionissues.

Mumbai, Maharashtra, In

Client: PROS Inc - US based Big Data Analytics CompanyResponsibilities:• Developed dataset processes for data modeling, mining and production.• Performed research, analysis and statistical modeling on complex data to execute pricing strategies.• Created models to forecast the price of a given destination is likely to go up or down using R and Python.• Interpreted the data to identify trends and suggested improvements to realize revenue growth using R.• Build and maintain SQL scripts, and complex queries for data analysis and extraction.• Created dashboards for server utilization and minimized system outages saving an estimated $1.2M annually.

Navi Mumbai, Maharashtra, In

Responsibilities:1) Development of a sustainable and modern framework for smartphone interaction with NFC enabled devices.2) Deploy and test the NFC framework in a controlled environment.3) Document performance and development effort.4) Present the project to a review panel.

Master'S Degree, Data Science

Bachelor Of Technology (B.Tech.), Information Technology

Computer Science