Infrastructure as Code
Terraform · Pulumi · Ansible · Crossplane
6y experiencetoday
DevOps Engineer
Navigating the infrastructure between stars.
I build systems that survive the long journey — resilient, observable, and quietly automatic. Currently charting deployments across the Kubernetes void.
MISSION LOG // 01
Seven years charting systems through production.
I started deploying cron jobs on a shared hosting account. Now I orchestrate 200-service Kubernetes clusters that handle millions of requests per second. The journey between those two points taught me that infrastructure is not about the tools — it's about the people who get paged at 3am.
I work in Rust, Go, and Python. I prefer boring technology that works. I write runbooks before I write code. I believe observability is a feature, not an afterthought.
200+
services orchestrated
99.97%
uptime maintained (last 18 months)
47
production incidents resolved
12,000+
deployments shipped
NAVIGATION CONSOLE // 02
Tools I trust to keep the ship flying.
Container Orchestration
Kubernetes · Docker · Helm · ArgoCD · Kustomize
5y experienceyesterday
CI/CD Pipelines
GitHub Actions · GitLab CI · Buildkite · Jenkins
7y experiencetoday
Observability
Prometheus · Grafana · OpenTelemetry · Datadog · Loki
4y experience2d ago
Languages
Go · Rust · Python · Bash · TypeScript
8y experiencetoday
TRAJECTORY // 03
Seven years of deployments, three companies, one throughline.
Vector Cloud
Staff DevOps Engineer
Sep 2023 — Present
- Led migration of monolith to 47 microservices on EKS, reducing deploy time from 47 minutes to 6 minutes
- Designed multi-region active-active Postgres setup with sub-200ms cross-region replication lag
- Built internal platform team of 5, shipping 12 internal developer tools adopted by 200+ engineers
AWSKubernetesTerraformGoDatadog
Drift Labs
Senior Site Reliability Engineer
Aug 2020 — Aug 2023
- Reduced P1 incident MTTR from 34 minutes to 8 minutes through SLO-driven runbook overhaul
- Migrated logging infrastructure from self-hosted ELK to OpenTelemetry + Grafana Cloud, saving $180k/year
- Authored incident response playbook adopted by 8 of 12 engineering teams company-wide
GCPKubernetesPrometheusPythonTerraform
Hooli
DevOps Engineer
Apr 2018 — Jul 2020
- Containerized 23 legacy Java services, cutting infrastructure cost by 41%
- Built CI/CD pipeline processing 800 builds/day across 14 repositories
- Mentored 3 junior engineers through internal SRE certification
AWSJenkinsDockerBashDatadog
WORMHOLE // 04
Open source that survived contact with production.
Pulsar
GitOps Deployment Visualizer
Real-time topology graph of K8s deployments across clusters. Catch failed rollouts 4 minutes before Prometheus alerts. Built for teams running 50+ services.
RustWebAssemblyD3.jsKubernetes API
1.2k
GitHub stars
89
contributors
340+
deployments
Terraform Drift Detector
Catch infrastructure drift before your next deploy
Continuous reconciliation that compares live cloud state against your Terraform plan, with PR comments on every diff.
GoAWS SDKTerraform Cloud API
K8s Cost Optimizer
Right-sizing recommendations with real workload data
Analyzes 30-day Prometheus metrics to suggest CPU/memory requests that match actual usage. Saves 20-40% on node costs.
PythonKubernetesPrometheus
TELEMETRY // 05
Reading the instruments, not the brochure.
GitHub Activity
847 contributions in the last year
Certifications
- CKAFeb 2023 → Feb 2026
- CKSMay 2024 → May 2027
- AWS SAPNov 2022 → Nov 2025
- TerraformAug 2023 → Aug 2025
Conference Talks
Mar 2024 · KubeCon EU 2024
Multi-Cluster GitOps at Scale
Sep 2023 · SRECon Asia 2023
Runbooks Are Code: Version-Controlled Incident Response
Aug 2023 · DevOpsDays Jakarta 2023
Building Internal Developer Platforms That Don't Suck
Open Source
4
CNCF projects maintained
847
merged PRs across 23 projects
12.4k
Stack Overflow reputation
TRANSMISSION // 06
Open channel for collaboration.
I'm available for Staff and Senior DevOps roles, consulting on Kubernetes migrations, and the occasional conference talk. Best response time is over email — I read everything within 24 hours.