Karthikeyan Sukumaran

AI App development with OpenAI/ML. Data modeling & transformation using DBT Cloud data infrastructure on AWS Scalable ETL/ELT pipelines with Python Data lake & warehouse architecture (Redshift, S3, Glue) CI/CD for data workflows Mentoring Engineering Teams

• Led many ETL/Cloud projects including modernization of legacy database and Data Migration from Netezza,DB2 to Snowflake using AWS,Datastage,Unix. • Designed and deployed many data solutions with ETL data pipelines and implemented the best practices in snowflake and ETL to improve performance and collaborated with different teams for data sharing. • Implemented many projects with IBM Datastage version upgrades involving cyberark along with data tokenization using protegrity. • Adept in creating and modifying existing reusable components like scripts and etl jobs to automate the functionality and improve the performance and efficiency. • Created resuable framework in Hadoop to create Data Lake to transfer data back and forth between many sources like mainframe, Databases to Hive and vice versa using sqoop in hdfs and unix with quality checks that validates the data. • Implementation of Modernizing data solutions involving Cloud Migration of legacy on-premise Databases with 100+ million records to Snowflake cloud via ETL,DBT,AWS for a top banking client. • Provided Data migration solutions involving 10+ years of huge volume of data in TB sizes in data lake from Hortonworks Hadoop platform to Cloudera Hadoop platform. • Analyzed and designed the ETL pipeline using ETL/Big Data technologies to accommodate business requirements using ETL DataStage, Snowflake, DBT, Hadoop,Hive,Python and Unix Scripting. • Created snowflake stage table from AWS S3 bucket and loaded data from stage tables to Snowflake table using the best practices for optimized performance. • Provide regular support guidance to ETL DataStage project teams on complex problems and issue resolution. • Experienced in working with structured data using Hive QL, SparkSQL, join operations, Hive partitions, bucketing and internal/external tables. • Utilized Sqoop Scripts to ingest data from different RDBMS sources into Hadoop cluster(HDFS) and created Hive tables, partitions, data loading into hive tables.

• Understand the business requirements for setting up the ETL/Cloud environment for integrating with internal and external source systems. • Created the design document and ETL pipeline based on it to migrate of large data sets to create foundation layer and cloud warhouse • Involved in development and testing to ensure and verify completeness of data, preventing data leakages due to migration on large data sets • Implementation of ETL jobs in production and further optimization of ETL jobs on the platform and Continuous monitoring of the data pipelines and system scalability • Developed technical documentation on ETL, migration and other query runs • Unloaded the data from databases like Netezza,Db2,Oracle and load the data into S3 bucket and process those files in cloud warehouse like Snowflake and data lakes like Hive/Hadoop and vice versa. • Created and troubleshooted Hive queries that helped business analysts to spot emerging trends by comparing current data with GPM reference tables to calculate metrics.

• Involved in requirements analysis and business logic based on provided documentation and work closely with tech leads and Business analysts in understanding the existing system. • Understood business requirement from Business to identify data coming from different sources and created source to target mapping to design Data pipeline. • Creation of Data lake in Hadoop/BigData using ETL and Unix scripting from various sources like Db2,Netezza,Oracle and mainframe/FTP/Windows servers.

Extensive Experience in ETL Develpment,transformation and migration using Ascential datastage 7.5, IBM Data stage 8.1 & 9.1 and PeopleSoft fin upgrades

• Development and Testing of ETL Jobs in Datastage as per business requirements with data coming from various sources from banking sectors. • Optimized the ETL jobs in Datastage jobs and Sql and improved the performance and efficiency of the job run in quick time without warnings. • Providing implementation support so that the target systems receive their inputs in accurate time without any delay. • Developed many reusable components in UNIX shell scripting for automating the data pipelines and data wrangling to handle data issues. • Extensive production on call support and experience with jobs tunings and handled time sensitive issues without any delays. • Scheduled Jobs in Control M tool and monitored for job failures.