Mario Leonilo III Marcos

• Introduced agile methodologies and development best practices to division to enhance product development.
• Served as subject matter expert on datasets, taxonomy mappings and model results for both clients and internal team members.
• Curated Datasets for model training using Pandas and Pyspark which included translating languages using a deep model (MarianMT).
• Automated and simplified curation process using in-built technology that uses Airflow to allow for new users to curate data.
• Ran Oracle DB Queries to gather needed datasets.
• Used Amazon AWS with Boto3 to store curated datasets.
• Investigated and identified issues for misclassified predictions (whether it originates from Model or Dataset).
• Created Tableau Reports (Sankey charts, Prediction drill downs) to present dataset findings and changes.
• Built working pipe-line with team that automates extraction and curation process for dataset.
• Worked and coordinated with other teams using Agile Methodology to integrate, maintain and deliver professional products.

- Designed and implemented an ETL pipeline for the
processing of medical, RX claims and other patient
information.
- Orchestrated a replication process via AWS Glue,
and Step functions facilitating seamless integration
between Postgres, MySQL, and Redshift tables.
- Created a data coherence monitoring tool in
Quicksight for alignment across Redshift, Postgres,
and MySQL tables.
- Assumed responsibility for the Redshift database,
serving as the main repository for processed data.
- Configured and managed AWS EC2 instances to
optimize infrastructure performance.
- Executed Terraform scripts to whitelist specific IP
addresses.

• Regression and Positive Testing.
• Onboarding and Reonboarding Actions.
• Test Strategies.
• Editing of JSON scripts to fix actions.
• Troubleshoot bugs.
• Use of ADB for Intent Deep Links.