Software Engineer
CurrentResponsible for keeping our bank integration infrastructure alive and providing tooling so that an ~80 engineer organization can stay on top of the health of their integrations. We support 12,000 institutions, investment platforms, and other financial providers in the US, Canada, and Europe. Each geography has unique uptime baselines, providing us a constant challenge of managing alert fatigue with quality and alert usefulness.When not wrangling PromQL queries to keep our integrations engineers happy, our team ensures our enormous Kubernetes clusters stay healthy, in conjunction with our primary infrastructure team. They own the Kubernetes infra, we own everything on top of that. Due to a number of complex factors, we frequently maintain state across pods and have spent countless hours wrangling obscure networking-level bugs. We also manage custom deployment logic for our largest ~6000 pod service and can deploy it dozens of times a day.I have owned significant reliability efforts that cut across our stack, constantly pushing for improved observability and service resiliency.I volunteered our team to be a guinea pig for Plaid's original migration to Kubernetes from ECS. We learned a lot, and realized it certainly takes a while to move large, business critical services between clustering methodologies. Especially with 0 tolerance for downtime. We survived to tell the tale.Other teams know me for being a significant culture carrier for both my team and the entire engineering org.Generous quotes from coworkers include: "Austin is also a serious driver of personality of the team.""He is very motivated to solve problems for other teams.""The first to praise engineers when they accomplish something especially difficult, which is not something to be taken for granted."SEO appeasement: NodeJS, Typescript, GoLang, ELK / Kibana, CloudFormation, plus everything listed previously, except Java.