↓Skip to main content
  1. Work Experience/

Amazon Web Services - Software Developer II

·2 mins
Table of Contents

I worked on two software teams at AWS, owning two different products.

AWS Comprehend Medical #

AWS Comprehend Medical at the time was a public AWS service offering NLP over medical documents (named entity recognition, relationship extraction).

I assumed leadership on a Java codebase migration effort with a group of 3. The migration successfully deprecated over 200K net lines of code, and produced indirect improvements on entity detection, latency, and throughput for the InferSNOMEDCT, InferRxNorm, and InferICD10CM APIs.

I designed and developed a lightweight, domain-specific validation tool that allowed us proceed with the migration without measurable regressions.

Separately, I led slices of implementation work for a new feature - detection of Behavioral, Environmental, and Social medical entities. While our team was onboarding newer members, I took point on getting requirements and feedback from product and ML teams and translating them into concrete engineering work.

I got very familiar with multi-threading and Java’s concurrency library.

AWS HealthOmics #

AWS HealthOmics at the time was a public AWS service providing genomic and transcriptomic data storage and analytics.

For the release of Annotation Store versions, I created designs for storage, wrote microservice logic, wrote test plans, and led threat modeling review with application security teams. Despite being a newer member, I reviewed every single piece of code contributing this feature, providing feedback and leading discussions across various layers.

After this, I led our team of 5 through design, implementation, and release to support table partitioning for Variant Store data. I got very familiar with tools in the big data and genomics ecosystems: Apache Spark, Apache Iceberg, genomics file formats (VCF, gVCF, tabix). Along the way, I led development for various reliability improvements for our asynchronous genomics file ingestion system.

My experience here gave me foundational first-hand knowledge on what is needed to architect large systems for scale and resilience. AWS services I became highly proficient in: AWS Glue, DynamoDB, S3, Lambda, SQS, Step Functions, IAM.