MOHAN D

1. Identification and Resolution of various Data Quality issues with TLD data warehouse (Teradata)

2. Data modelling & Development of ETLT utilizing Informatica Mappings & Teradata bteqs (SQLs)

3. Enhancement of existing ETLT process for performance tunings and to resolve data quality issues

4. Data remodeling and development of ETL process using HiveQL in Horton cluster

• In Epocrates – Data Project, the App’s tracking data & purchase data is loaded into EDW and data is copied into SAS datasets by SAS procedures.
• By utilizing EDW data, OBIEE & SAS reports give various business insights such as advertisement impressions & clicks, product usage, user details, application behavior, user behavior, application bugs, revenue details and more.
• These insights are primarily used for targeting the advertisements to specific group of users, for product enhancements and improvements, to calculate & forecast the revenue, to increase application adaptation, to gain more revenue and for other important business decisions.
• In Epoc 2.0, the goal is to remodel the application and to move to Service Oriented Architecture.
• Key Activities:.
• Have worked on Oracle SQL for complex report requirements utilizing the EDW data.
• Have worked closely with the stakeholders on several data and report issues, financial data issues and have fixed them on time.
• Have done data-modelling for some of the sub-products of Epocrates and have improved the existing data models working with DBA to increase performance and stability.
• Developed Informatica mappings & workflows and developed OBIEE reports for new requirements.
• Have improved performance and stability of the existing Informatica workflows.
• Have automated audit process using shell scripts for the existing jobs to proactively prevent data quality issues in EDW and in SAS datasets.
• Have found data issues in application tracking logs and proposed solutions to respective teams to rectify the same.
• Effectively led a team of size 3, taken care of team’s tasks and worked towards achieving the goals as a team in Athenahealth Pvt.
• Have developed Spark-Streaming application using Scala to perform near real time ETL into Cassandra by consuming Kafka messages for Epoc 2.0.
• Have done data-modeling in Cassandra based on the report requirements and modified the existing Spark-Streaming applications based on the new model for Epoc 2.0.
• This has significantly improved the performance of the Auto-Refresh dashboards in Zeppelin.
• Have used Sqoop to import data from Oracle and developed Spark ETL using Scala to load the data into Snowflake for Epoc 2.0.

• The entire process is automated through Shell Scripts.
• Some of the source data is extracted from Oracle using SqlPlus & Informatica and some aggregated data is taken from Teradata warehouse and sent into downstream/client companies through FTP, SFTP.
• Based on the viewership and calculated ratings data, NBC will make business decisions for example deciding the price for advertisements, deciding the shows that to be continued and the shows that are to be cancelled.
• Key Activities:.
• Participated in Data Modeling, Informatica Code development for AMRLD module for 40+ dimension and facts in between Jun 2014 and Aug 2015.
• Developing Teradata BTEQs (SQLs) to load on SCD (Type 2) and SCD (Type 1) Dimensions and also to load fact tables which involves complex calculations and logic for AMRLD module.
• Developed Teradata SQL to produce complex reports.
• These reports shows good insights into the viewership data for any telecast in minute level.
• Developed Informatica mappings, workflows and Teradata bteqs and performed code changes during different phases of development process.
• Also automated the entire ETL/ELT/ETLT process using Shell Scripts.
• Have done end-to-end development for a module called ‘Rentrak’.
• Done Performance tuning for LMW module utilizing Informatica PowerCentre and Teradata SQL knowledge and obtained 72% faster load process.
• Appreciated by the clients for the same.
• Utilized in-depth knowledge in Shell Scripts (UNIX) and automated several Audit and Validation Processes and have reduced manual effort by 95%.
• Have worked on the Production Issues and Outages that arise during ETLT process or in the generated reports.
• Have worked on process improvement activities.