Radu G.

I worked on an IoT project, where measurements data from pools are read from multiple transactional databases into a Delta Lake in Azure. The Databricks Delta Lake contains multiple Databricks notebooks that ingest data from transactional databases and processes the data to output information used in PowerBI reports. I build ETL pipelines with PySpark, create the data sources needed for the PowerBI reports and also help with creating and deploying the PowerBI reports in the online service.

Worked as a Data Architect to create the data model, ETL processes and outlined the documentation template for migrating data stored in an old DBASE database to SQL Server 2016. Created custom T-SQL Scripts and migration pipelines automated with Azure DevOps. I outlined the general data model, set up documentation best practices using Enterprise Architect, created T-SQL templates that could be called from SQLCMD, templates which could be "injected" with specific table-based migration logic.

Worked as a Data Scientist on a greenfield project where I had to predict errors in a factory production line. The purpose was to create a machine learning algorithm for predictive maintenance so that a production line could review the state of a certain step in the production line, so that fewer faulty products reach the end of the production line. The algorithm would predict the likelihood of an error occurring in a certain step of the production line and raise an alarm to the on-duty person

Data conversion and migration from older systems. Refactoring and updating SQL code to use newer functionalities. 1 on 1 discussions with developers as well as organizing trainings on querying best practices, performance tuning, index tuning, set-based thinking and other best practices. Identifed and improved performance for low performing queries through query rewriting, index tuning. Administrator tasks involving backups and automating restores via Powershell scripts.

Review and assist Java developers with writing Oracle PLSQL scripts. Worked on improving query performance for complex queries, solved issues with Goldengate replication. Actively tried to improve the overall PLSQL knowledge, relational and set-based thinking in the teams of developers through 1:1 discussions with developers and technical presentations which focus on explaining different database concepts like database/table design, indexing strategies, querying best practices.