Counting proteins by hand

Keeping the company connected by architecting compute+data solutions and the engineering team to build them. Using Kubernetes, Bazel, JAX, S3, RDS, and Python.- Architected and built core infrastructure for ML training- Hired, led, and managed several engineers on my team, and mentored several more- Championed and modeled a culture of caring, technically and socially, and of elevating others across the company- Created and maintained multi-cloud Kubernetes clusters (EKS/GKE/self-managed) running both CPU and GPU accelerated workloads- Planned, acquired, racked, and managed on-prem/colo networking and GPU servers- Pioneered adoption of technologies including Pulumi and Terraform, NixOS, and Tailscale

Technical Lead in Ground Truth, internal tools (primarily Atlas: https://www.wired.com/2014/12/google-maps-ground-truth/) for visualization, curation, and evaluation of Google's map data- Led the vision and technical direction for my 7 person team transitioning from a Java Swingdesktop application to a TypeScript/WebGL web application with a Java backend- Concurrently led a 3 person team incrementally rearchitecting an AngularJS app optimizingfor maintainability- Worked with key stakeholders to drive prioritization and investment in our products- Designed and implemented schema for modeling map features relevant to automakers

TL for Atlas, enabling visualization and curation of Google's map data.Previously: focused on G Suite customer onboarding systems (marketing content, signup flows, domain registrations, and DNS management) in order to optimize growth.

Developing Eclipse plugins to make research software easier to use.

Developed an internal GWT application to enable engineers to debug broken data in a proprietary production system.