• As part of the Personalized AI Capabilities team at Meta Reality Labs, developed features that added individualized context to AI interactions on wearable devices.

• Contributed to launch of Reminders feature for Ray-Ban Meta smart glasses, announced at Meta Connect 2024, working with Kotlin on the Android version of Meta Wearables App.

• Implemented features around synching reminders and associated media, providing a consistent experience across devices and when reinstalling.

• Added metrics logging for reminders operations to track usage analytics, end-to-end latency, and error rates.

• Integrated user contact embeddings to refine personalized voice recognition when searching for contacts, enhancing accuracy and ease of use.

• Member of Common Tools and Services team building and maintaining internal applications to support other teams.

• Wrote Ruby on Rails service that takes parameters from a React UI or GraphQL API and turns them into a group of configured Kubernetes objects providing users easy access to hosting, DNS, monitoring, logging, Vault secrets integration, and rollbacks, eventually responsible for running nearly a thousand user services.

• Contributed to existing Rails project for managing and monitoring cronjob tasks, including improving page load times via PostgreSQL query optimizations and stability bugfixes.

• Created Python application for the automated, scheduled, non-disruptive termination of Kubernetes nodes, which later was revisited to tailor to other teams' unique use cases and for cycling out old nodes during cluster migrations.

• Migrated a legacy in-house user login Java service to Okta, providing additional security to users and reducing oncall tickets by roughly 10 per week.

• Upgraded Kubernetes cluster versions five times, including both migrations to separate clusters where services had to be moved by me with a seamless switchover as well as Amazon EKS in-place upgrades, every time requiring proactive communication with other teams about API deprecations and deployment windows resulting in little-to-no downtime.

• Maintained installations of 3rd party and vendored services, including Github Enterprise, Graylog, Sourcegraph, ArgoCD, Istio, among others.

• Worked on Ads Experiments Infrastructure team to improve internal tools.

• Updated existing internal system using C++ to reduce memory and resource usage, reducing the size of objects in memory by 75-96%.

• Created and maintained documentation and unit tests.