Alvin B.

Led data acquisition team of 4 engineers to develop and maintain data ingestion pipelines of Accolade.

Built on top of PySpark, AWS Batch, and AWS EMR.

Provided domain expertise to the task force responsible for integrating HealthReveal within Accolade.

Build custom data ingest process for EHR and Insurance healthcare data using Spark, Scala & Cassandra.

Introduced Snowflake at HealthReveal and migrated all data from Cassandra into a newly designed operational and analytical data model.

Refactored data ingest into a Python & Snowflake-based data pipeline.

Develop secure, external facing APIs in Kotlin & SpringBoot for client access.

Develop a data pipeline to consume HL7 messages using Java, Kafka, and Spring Boot.

Leverage ElasticSearch & fuzzy logic matching to build an algorithm to handle Person mastering across different sources like hospitals, nursing homes, and insurance rosters.

Designed and built internal Patient Roster loading and validation tools to enhance internal efficiency.

Worked on React frontend to build custom components to display flags identifying high utilizers, high-risk patients, and length of stay, among other analytics.

Develop a data pipeline to consume HL7 messages using Java, Kafka, and Spring Boot.

Leverage ElasticSearch & fuzzy logic matching to build an algorithm to handle Person mastering across different sources like hospitals, nursing homes, and insurance rosters.

Designed and built internal Patient Roster loading and validation tools to enhance internal efficiency.

Worked on React frontend to build custom components to display flags identifying high utilizers, high-risk patients, and length of stay, among other analytics.

Built a centralized validation framework to serve front-end and back-end account data tools.

As the team lead, designed and implemented a dynamic client onboarding portal using Drools, Angular, and Java.

Hands-on Team Lead of the Document Rules Engine, in charge of 4 developers tasked to build a rules-based document requirements calculator to serve multiple units across Citi.

Built an API service on top of the internal ESPN CMS system.

Used Flash to implement video engagement tracking features.

Implement cue points based on in-video ads display.

Built Java-based loader framework to process end-of-day pricing data files for reconciliation.

Built a Java Swing desktop app for account management.

Implement business validation rules as a standalone microservice to be re-used between the desktop and loader apps to ensure consistency and standard error messages.