Poojaanjali Thillai

Leading data engineering initiatives for automotive clients (BMW, Daimler Trucks), spanning cloud ETL pipeline development, infrastructure automation, and applied GenAI innovation. Architecture Solutioning (Analytics Pilot) Authored the high-level target architecture blueprint for a SAP-to-Power BI analytics pilot on Databricks — defining a four-stage pipeline (SAP S/4HANA or ECC → Azure Data Factory → Databricks → Power BI) for a greenfield rebuild of a representative Sales Dashboard use case, rather than

• Used Prefect Scheduler to pull data from Redshift and perform transformations in Python to load into S3. • Deployed Prefect orchestration using an AWS EC2 instance and Prefect server. • Used Amazon Redshift Spectrum to analyse Redshift data and AWS Glue to create/match data schemas. • Migrated data between Redshift clusters using Redshift Spectrum and S3. • Built a pipeline for different business units using Kinesis Firehose, S3, Lambda, and Spectrum. • Wrote unit test cases to validate additional code im

• Programmed Pyspark Scripts to process and load the sql/avro/json files into hive. • Scheduled the Pyspark scripts using oozie. • Programmed Spark Sql scripts to extract and transform the data in hive and impala as per business requirements stated by the stake holders and users. • Scheduled the Pyspark scripts using oozie. • Possess good business knowledge through constant interactions with stakeholders to code accordingly. • Used Egress to establish connection between Pod and external network. • Performed

• Helped client by suggesting strategic solutions by creating an interactive Shiny Visualisation dashboard and Implemented Machine Learning Algorithm on Demographic and COVID-19 datasets. • Created a Primary Care Audit Excel sheet using macros, Data validation, formulas etc.

Worked on Large Structured Dataset in a systematic approach using Hadoop, Data Warehouse, ETL tools to give the correct data to Data Scientist team. •Used Spark for faster data processing and for reading, writing data on Hive and Hbase(NoSQL). •Have given Knowledge transfer to team members on Spark and documented on the same. •Created and Used Python UDF’s in hive. •Used Unix Scripting for Automation of Data Pipeline process. •Scheduled and monitored the Hadoop and Unix Production jobs using Oozie and Run-d

Key Modules- R, Python, Java, SAS, Spatial Database, Geo-Graphical information, Linear Modelling, Statistical Machine Learning, Case Studies on Data Science

Key Modules- Fundamentals Of Computing, Data Structures and OOPS, C and C++