• Building and encouraging platforms around Youth First!• ETL, Warehousing, BI visualization (Looker Studio, Tableau, PowerBI)• Designing, administrating, and programming with databases

• Apart from secondary NoSQL databases (MongoDB, Redis), we were using a MySQL variant from Percona as the core and transactional database: It was a combination of Percona XtraDB Cluster and traditional MySQL Replication. The cluster was also load-balanced and secured by the application gateway ProxySQL.• I monitored the database health with the help of Grafana-based dashboards and solved incidents. I was the center of contact for every problem related to the MySQL system. There was only me as a database man under the Engineering Director. • We were operating an on-premise data center for Production while for developments, we used the AWS Cloud. However, we didn’t subscribe to AWS services (neither MySQL RDS nor Aurora) but kept using our old database solution in EC2 instances.

• In PALTECH, I worked under the Data Director who sat remotely in Singapore. I was in charge of most activities in Vietnam and the contact point for BA, Developer, and QA/QC teams.• I was a Database Developer for a MySQL system at first:I easily applied my database knowledge with Oracle into programming with MySQL database. I wrote MySQL stored routines, designed and built the entire reporting database.• I learned more about MySQL internals and also joined the database administration.

• I took part in customization of a core banking system named Oracle Flexcube:The job required coding inside the Oracle database in PL/SQL language. It was quite amazing that the application’s business logic was implemented in PL/SQL executed inside the Oracle database. On the other hand, the application UI was developed in Oracle Forms technology, and I also developed Oracle Forms modules.Programming with Oracle database truly made the initial steps in my transactional and high concurrency design.• In the bank, I encountered the problem of disintegrated systems: There were distinct sources of data, making data access troublesome and also creating divergences. The bank needed a single source of truth as well as less reliance on the IT support for data extraction and reporting activities. The choice was to build a warehouse to integrate data from diverse sources. Also, a banker could create ad-hoc queries by dragging and dropping GUI items in an interactive analytical tool. Therefore, I had the chance to equip myself with Data Warehousing techniques.• In the project to upgrade Oracle Flexcube to its latest version, I was trusted to research oncustomization in the new system:The Java Enterprise replaced Oracle Forms, and JavaScript was used extensively at the frontend. It was still awesome that the entire core logic was kept in PL/SQL code. In other words, only the outer shell of the application was replaced. Just to say how strong and mature the Oracle database language was!