wwest•dev

SRE lead with a goal of putting the infrastructure in a more stable position to deliver value and attract further investment in a tough market: set strategic goals for infra; authored 2-year infra roadmap, did much of the initial execution in the first 2 quarters:

• Update legacy hand-built GCP infra w/ normalized infra subject to a Terraform-based workflow. Introduced containerized workloads.
• Replaced hand-managed config + secrets management with modern hierarchical/environment-aware solution based on Ansible + a modern secrets provider.
• Automate SRE oncall toil, especially that required by industry compliance, without cutting corners.
• Replaced outdated monitoring (Zabbix) with a cloud-native stack (Prometheus/Grafana) with robust data collection, calculation, reporting, and alerting capabilities, established SLOs based on Apdex and drove down error/alert noise to 1-2 pages a week while measurably increasing reliability.
• Maintained technical PCI compliance for a large-scale POS system, including infrastructure and product code changes.
• Embarked on R&D-wide effort to level up devs, SREs and support staff on operations, log/code/metrics hygiene, oncall discipline, post-mortems, and fire-drilling/role-playing incident scenarios and live debugging techniques.

Develop & manage LinkedIn Terraform-based Infrastructure-as-Code (IaC) platform. Developed and drove effort to successfully operationalize the MVP using prior experience as SRE, as well as deep SWE skills e.g. development of thread-safe metrics emission client in Go, expansion of API and CLI authz, SSO in Python.

Custom provider development and provider/IaC developer support. Drove partnership with Kubernetes team to support pod declaration in IaC and implemented custom authentication, provider and environmental support for K8s use cases.

Designed hands-on conference workshop to train existing and prospective users in IaC development and platform use.

Most recently worked on greenfield policy-as-code platform (OPA-based), first as a sidecar to the IaC service, and then as a general service.

Two years embedded in the product team for LinkedIn Recruiter. High-revenue, fast-moving environment requiring fast triage skills, and ability to collaborate in product codebases to provide improved operability; JMX management, cache loading + validation with Apache Gobblin. Wrote internal productivity tools for peers (e.g. Ember wrapper of JIRA API, hadoop flow monitoring tools). Maintained end-to-end tests of the Recruiter platform.

Two years as a capacity engineer, working on live performance testing of production apps, capacity estimation and planning, and private cloud resource efficiency. Extensive work on REST APIs and documentation, feature development for perf testing framework, planning and development of core operational efficiency metrics.

Primary architect of and full-time contributor to WebMD's infrastructure provisioning automation framework, including public cloud provisioning and Chef bootstrap automation.

Primary author of a configuration data management system, built on Rails & Neo4j and integrated with Chef. Author and maintainer of several smaller support tools, cookbooks, test suites, and documentation.

Led a dev team maintaining Yale's ServiceNow instances. Transformed team structure and practice away from Waterfall and toward agile methodologies, ultimately delivering small batches weekly and driving the introduction of automated feature testing.

Original role as UNIX systems admin in a central UNIX group serving many customers with a complex ecosystem of academic and business applications, databases, and web services. This included Linux and AIX support for a variety of databases & applications (Oracle, LAMP, Java, Rails, Atlassian stack).