• Maintains, extends, and serves as subject matter expert for 10+ microservices within the MyFitnessPal backend ecosystem that power the app's core diary, food logging, and AI logging experiences

• Partnered with Product and Data Science teams to design, integrate, and revamp the app's AI logging experience, which drove incremental increase in feature engagement and revenue

• Modernized the AI logging technical stack, which enabled observability into LLM workflows and supported dataset-driven product decisions through Langsmith integration

• Successfully drove a platform modernization of a high-traffic and highly visible diary and logging microservice delivering:

• Zero downtime to customers throughout the migration

• 40-70% latency improvements for core application endpoints

• 50% improvement in developer velocity

• Migration off of end-of-life (EOL) technologies

• Architected and executed the migration of Kafka logging events off legacy infrastructure dependent on EOL technologies

• Participates in on-call rotation, responsible for incident triage and response across 100+ microservices in the backend MyFitnessPal ecosystem

While a core skill of software engineering is the ability to quickly learn and adapt to new technologies, I have particular experience with the following:

• Languages: Python, Kotlin, SQL, Ruby, JavaScript, TypeScript

• Runtimes, Libraries & Frameworks: FastAPI, Spring, Node.js, Express, Rails

• Cloud, Infrastructure & Observability: AWS, Kafka, Docker, Redis, Datadog, Kubernetes, Planetscale, PagerDuty

• Misc. Technologies and Skills: LLMs, VLMs, Langchain, Langsmith, REST APIs, Linux, MySQL

Twine is an open-source drop-in WebSocket infrastructure for modern web applications that provides connection state recovery. Twine drops into an existing application via client and server libraries and is automatically deployed into a user’s AWS account.

• Architected and constructed a drop-in auto-scaling WebSocket infrastructure hosted on AWS (EC2, DynamoDB, Elasticache for Redis, Elastic Beanstalk)

• Load-tested 96,000 concurrent virtual users over 20 minutes, successfully

• Built out the backend REST API (Node.js, DynamoDB, Redis) including constructing a decoupled Redis pub/sub mechanism for sending data received from backend servers out to clients via a WebSocket connection

• Authored comprehensive technical case study, readable at: https://twine-realtime.github.io/

• Collaborated remotely with an international team of four engineers across three time zones using agile workflow

• Technologies: JavaScript/TypeScript, Node.js, Express, WebSocket, Socket.io, Redis, DynamoDB, AWS, Docker

Head & Neck Medicine

• Clinical nurse caring for critically ill oncology patients

• Preceptor to new graduate RNs

• Selected to be trained as a discharge coordinator to participate in a hospital-wide initiative for improved patient flow and experience