Led Zillow's transition from a data lake architecture to a modern data stack. Delivered a self-service data platform that has enabled over 140 teams across Zillow to successfully onboard new data workloads in minutes using a unified, cloud-native control plane developed in Kubernetes. Teams have successfully deployed over ten thousand resources including Kafka topics, Kafka Connect Jobs, Flink Applications, and event ingestion pipelines. Enabled decentralized ownership of data products through streamlined onboarding processes, high-quality documentation, automatic data observability, and federated data governance.Led the successful adoption of dbt within Zillow’s data organization to dramatically streamline and simplify data pipeline development. 18 different teams at Zillow have adopted dbt and have used it to deliver 2000+ new models to production.

- Designed and delivered a self-service, fully managed Flink platform enabling any engineer at Zillow to bootstrap and deploy production-ready Flink applications to Kubernetes with out-of-box monitoring, CI/CD, and diagnostics tools. This work enabled 24 different teams to rapidly adopt Flink for stateful stream processing with little or no prior knowledge. These Flink applications drive mission-critical product features across the company with 24/7 uptime, including home listings data and personalized experiences.
- Designed and delivered a resource provisioning system backed by a custom Kubernetes operator enabling any engineer at Zillow to quickly configure and provision Kafka topics, ACLs, and many other resource types on a shared, fully-managed infrastructure.- These two systems combine seamlessly to simplify all aspects of developing and running highly-available, event-driven services in production so Zillow engineers can focus on delighting their customers. As of May 2021, 60 different engineering teams at Zillow have provisioned over 1300 Kafka topics and 300 Flink clusters, all backed by a shared, fully-managed infrastructure running on thousands of nodes.

- Served as a technical project lead, designing and delivering a near-real-time distributed tracing platform for tracking and measuring the E2E performance of the Windows build, media creation, and test workflow. Took a lead role in designing a distributed microservice architecture for stitching together millions of telemetry events per day into a directed acyclic execution graph and using it to generate business intelligence in near-real-time. Promoted to Senior Software Engineer (Aug '18) based on my leadership contributions towards the successful delivery of this platform. A patent was internally filed for our graph generation approach.- Delivered and maintained a 24/7 distributed cloud service. Individually implemented telemetry, distributed tracing, metrics collection, and live-site monitoring for three different microservices. Championed continuous integration and deployment practices within my team ensuring that all microservices and SDK packages could be automatically deployed, tested, and validated. Technologies used: C#, Azure CosmosDB, Redis Cache, Apache Spark, Azure Data Lake, Service Fabric, Azure Functions, Stream Analytics, Angular, SignalR- Served as a platform evangelist within the organization. Drove platform adoption with 10 different partner teams by giving presentations to leadership, consulting with service owners, and delivering an SDK and accompanying documentation. Collected requirements from partner teams, designed features, and oversaw their execution and delivery by our engineering team.

- Designed, secured funding for, and oversaw the construction of the Windows build system’s first comprehensive data model and pipeline to deliver key insights to engineering system stakeholders.- Received a promotion (Mar ‘15) for delivering a distributed build technology and improving compile times by 60%. Received a special performance award (Jan ’16) for leveraging data insights to improve distributed build throughput by 160% at no cost; the equivalent of $32 million cost savings.- Established a long-term partnership with a data science team within Microsoft and acted as the primary point of contact, discovering insights and enabling tactical, sustainable improvements to decrease compile times by over 30%. Leveraged data science expertise to direct investigations of complex data sets including presentations to leadership.

- Received first place at the ACM SIGCSE undergraduate student research competition in Raleigh, North Carolina, March 2012.

- Worked closely with a JavaScript expert to research, define requirements, and build an open source API extension to Dojo Toolkit for accessing sensory data on mobile devices.