Bhaskar S Email & Phone Number

Columbus, Ohio Metropolitan Area

Built a NLPCoP product (NLP community of practice), groups of AI enthusiasts who share a concern or passion for NLP using AI and learn how to do it better as they interact regularly. Designed Model for Risk Audit and Inspection Category Reclassification to evaluate the possibility of observations into different categories, classify the observational data using category definition.Designed Model for Semantic Textual Similarity to Finding the similar deviation based on the historical deviations.Designed Model for Clinical Trial LPLV-DBL Prediction for different Phases to predict the time duration from LPLV to DBL for a clinical trial.Implemented Apache Airflow for authoring, scheduling, and monitoring Data Pipelines.Scheduled Airflow DAGs to run multiple Hive, which independently run with time and data availability.Involved in optimizing Snowflake cost optimization initiatives with data models and efficient queries.Proficient in utilizing Apache Spark within Databricks for large-scale data processing.Worked on Snowflake environment to remove redundancy and load real time data from various data sources into HDFS using Kafka.Developed Python-based API (RESTful Web Service) to track revenue and perform revenue analysis.Worked with Tableau for generating reports and created Tableau dashboards, pie charts, and heat maps according to the business requirements.Designed and Developed Scala workflows for data pull from cloud-based systems and applying transformations on it.Involved in handling large datasets using Partitions, Spark in-memory capabilities, Broadcasts in Spark, Effective & efficient Joins, Transformations and other during ingestion process itself.Designed Model for NLP Knowledge discovery to get the insights from YOF response data to understand the diversity of responses across various countries/roles, key actionable sentiments & themes.Context based mining of each questions on Context/Semantic based sentiment analysis using BERT.

Columbus, Ohio Metropolitan Area

Built KNN and XGBoost model for large customer to get the opportunity of growth (head room) and likelihood to adopt the feature.Worked on Kafka REST API to collect and load the data on Hadoop file system and used Sqoop to load the data from relational databases.Monitored and optimized Azure Data Factory pipelines for performance, reliability, and cost-efficiency, utilizing Azure Monitor and other tools.Created and modified several database objects such as Tables, Views, Indexes, Constraints, Stored procedures, Packages, Functions and Triggers using SQL and PL/SQL.Created HIVE Queries to process large sets of structured, semi-structured and unstructured data and store in Managed and External tables.Built recommendation engine to recommend office products to the customer, 26% growth has been seen in usage for the customers exposed to the recommendation using SVD, Matrix factorization.Built a regression model to identify customer whose monthly usage is about to change by +/-10% for Microsoft. Many models based on customer size and usage pattern have been built.Accomplished a 20% improvement in the error rates of each job time leading to improved customer satisfaction and reduction in operations cost. Data gathering and preprocessing to remove corrupt entries. Segmented the customers based on the usage, profiling, revenue using Clustering and launched offers basis on the business requirement, Data modelling on various techniques using CART, Random Forest, Classification techniques and custom-er retention using Cross-sell techniques.Principle Component Analysis (PCA) is done to reduce the dimensionality in the dataset. Built a Supervised Random Forest Model to detect fraudsters.Built predictive models via Logistic Regression, SVM & Random Forests to identify high risk customers and identify churn customer.Key Achievement: Designed the logistic regression model with AUC score of 0.88, Sensitivity 0.8, Specificity 0.83 & Accuracy 0.82

Alpharetta, Georgia, United States

Design and develop data and reporting solutions using Microsoft Azure Cloud and Power BI Reporting platform. Develop Architecture Design and Project Road-Map for legacy reporting Platforms to Cloud modernization using Microsoft Azure Cloud offerings – Azure Data Lake, Azure Data Factory, Azure SQL and Azure Synapse. Develop Complex Power BI Reports with transformations in Data layer and complex DAX developments to Support visuals for Productivity reports, Dashboard reports. Designed and develop Azure Data Factory solutions to process and load JSON files to Dimensions and facts for Power BI Re-porting solutions. Design Dimension modeling for Reporting requirements. Support On-Prem Data warehouse and ETL solutions running in SQL Server & SSIS for daily data loads. Optimizing and maintaining everyday data loads from discrete sources in on-prem as well as in Azure Cloud. End-to-end solution for automating DB deployments using Redgate SCA & Power Shell scripts for integrating with Jenkins and XL release. Design Data Modeling solution for Regulatory Reporting Requirements that has Multi-Currency Data requirements with Security Transactions in Financial Markets. Design and develop data and Regulatory reporting for Deutsche Bank. Designing and implementing data model for SEG, SEC & SEQ reporting requirements. Developing ETL solutions using SSIS for everyday data load into report tables. Developing various reports for all 3 regulatory lines with multi-currency and USD-EQ reports. Design and develop multi-currency Summary reports with previous day balance aggregations for different line items with multi row aggregations for different segments like Liquid Funds, MV Securities, LSOC and COTC. Design and develop SIDR report with column level aggregation for multi-currency and USD-EQ reports. Performance tuning everyday load, indexing and partitioning.

New York, United States

Leading the development and delivering SQL Server DB Migration from 2012 to 2014. Data masking the criminal data to bring the DB backup to offshore development and QA environments. Developing Data migration plan for data and objects and performing dry runs before cut-off. Developing initial migration plan of database using full and differential database backup and restore strategy. Developing strategy for handling the delta during the cut-off day and migrating delta from application data during DB Migration. Redesigning of Application log-architecture for better performance by changing the storage mechanism using partitions. Identifying and rectifying the SQL Server performance at server level and stored procedure / Query level. Engaged in client work environment directly to resolve the performance issues during the system peak times. Created Hive tables, loaded, and analyzed data using Hive scripts, implementing partitioning, dynamic partitions, and buckets.Involved in continuous Integration of application using Jenkins.Collaborated with infrastructure, network, database, application, and business analyst teams to ensure data quality and availability.Created Kafka producer APIs to stream live JSON data into multiple Kafka topics.Developed Spark Streaming applications to process data from Kafka topics and load the processed streams into Snowflake.Leveraged Spark's in-memory capabilities to handle large datasets efficiently.Utilized Spark's broadcast variables, joins, transformations, and other features for effective data processing.Experienced in working with EMR clusters and S3 in the AWS cloud environment.Troubleshot Spark applications to enhance error tolerance and reliability.Optimized Spark applications/jobs to improve efficiency and reduce overall processing time.Designed and implemented real-time analytics solutions using GCP tools such as BigQuery Streaming and Dataflow for rapid insights.

India

Extensive data analysis by writing Complex Queries and Stored procedures to identify the Stock mismatch for Inventory items Analyze and report the inventory data for stock tracking for Lot and serial and non-control items to match it with transac-tion data. Writing SPs and Complex Queries to adjust the Finance postings and to resolve stock mismatches. Setting traps in product in purchase order and GR screen to resolve mismatches inventory and goods and sales tables that lead to mismatch in Item reports. Designed a multi-user Access database application for project and time tracking.Converted, cleaned, analyzed, organized, and documented large datasets.Analyzed Medicare discharge profiles for senior populations in selected markets.Applied advanced mapping and presentation techniques to analyze demographic data and managed care penetration.Worked with business and QA team for their needs and developed action plans. Gathered requirements from business clients and developed visualizations using Tableau. Used Tabular model to provide more security to data at Row- levelUpdated and created new stored procedures and views in SQL.Monitored database resources and activities for SQL Server.