I build the platforms other engineers build on. Ten years across infrastructure and developer tooling, most recently cutting deploy times from 40 minutes to under 4 for a 300-engineer org.

Impact

40m → 4m

Deploy time

Rebuilt CI topology and cache layer

$1.4M

Annual cloud savings

Right-sizing plus spot migration

99.98%

Platform uptime

Sustained across 18 months

300+

Engineers served

Internal developer platform

About

I care about the gap between 'it works on my machine' and 'it works for three hundred engineers on a Friday afternoon'. Most of my work is unglamorous: build caching, golden paths, paved roads, and the kind of observability that means nobody has to guess at 3am. I have shipped internal developer platforms at two scale-ups and one enterprise, and I have opinions about all three.

Experience

Lumen Payments

2022 — Present

Staff Platform Engineer · San Francisco, CA

  • Led the migration of 140 services from bespoke Jenkins pipelines to a shared Tekton platform, cutting median deploy time from 40 minutes to 3m 50s.
  • Designed the multi-tenant Kubernetes topology now running production traffic across four regions.
  • Built the internal CLI that 300+ engineers use daily; adoption reached 94% without a mandate.

Harbor Commerce

2019 — 2022

Senior Site Reliability Engineer · San Francisco, CA

  • Owned the observability stack — Prometheus, Thanos, and a Grafana estate covering 2,000 services.
  • Cut alert volume 71% by replacing threshold alerts with SLO burn-rate alerting, which measurably reduced on-call burnout.
  • Ran incident command for Tier-1 outages and rewrote the postmortem process around blameless review.

Beacon Data

2016 — 2019

Infrastructure Engineer · Remote

  • Automated cluster provisioning with Terraform and Ansible across AWS, GCP, and Azure.
  • Contributed to the Beacon Cloud control plane, focusing on rolling-upgrade safety.

Skills

Infrastructure

KubernetesTerraformAWSGCPPulumi

Languages

GoPythonTypeScriptRustBash

Observability

PrometheusGrafanaOpenTelemetryThanos

Practices

SREPlatform EngineeringIncident CommandSLOs

Projects

kubecost-lite

Open-source cost attribution for Kubernetes namespaces without the enterprise price tag. 2.4k stars, used in production by around 60 companies.

GoKubernetesOSS

paved

An opinionated scaffolding CLI that generates a service with CI, observability, and deploy config already wired up. Internal tool, later open-sourced.

GoDeveloper Experience

slo-burn

A Prometheus rule generator that turns a plain-English SLO into multi-window burn-rate alerts. Grew out of the Harbor Commerce alerting rewrite.

PrometheusPythonSRE

Education

University of California, Berkeley

MS Computer Science — Distributed Systems — Distinction

2014 — 2016

Georgia Institute of Technology

BS Computer Science — Highest Honors

2010 — 2014

Certifications

Certified Kubernetes Administrator

CNCF

2023

AWS Solutions Architect — Professional

Amazon Web Services

2022

HashiCorp Certified: Terraform Associate

HashiCorp

2021
All templates