{
  "$schema": "https://raw.githubusercontent.com/jsonresume/resume-schema/v1.0.0/schema.json",
  "basics": {
    "name": "Dan Arbaugh",
    "label": "Staff Software Engineer",
    "email": "dan@danarbaugh.com",
    "url": "https://danarbaugh.com",
    "summary": "Staff-level engineer with 12+ years of experience and 7+ years building and operating multi-cloud infrastructure at O'Reilly Media. I design and lead the systems that provision thousands of ephemeral cloud environments per week across Google Cloud Platform (GCP), Amazon Web Services (AWS), and Microsoft Azure, managing capacity planning, resource lifecycle, and cost efficiency across a multi-tenant platform. Built on Google Kubernetes Engine (GKE) with Terraform for IaC, Datadog for observability, and async Python for workflow orchestration. Google Cloud certified. Open source contributor in Go.",
    "profiles": [
      {
        "network": "LinkedIn",
        "username": "danarbaugh",
        "url": "https://linkedin.com/in/danarbaugh"
      },
      {
        "network": "GitHub",
        "username": "danarbaugh",
        "url": "https://github.com/danarbaugh"
      }
    ]
  },
  "work": [
    {
      "name": "O'Reilly Media, Inc.",
      "position": "Senior Software Engineer",
      "url": "https://www.oreilly.com",
      "startDate": "2018-12",
      "summary": "Cloud infrastructure platform engineering, capacity planning, multi-cloud orchestration, observability, and AI/LLM learning environment development. Promoted to Senior Software Engineer in April 2023.",
      "highlights": [
        "Designed and built Cloud Labs provisioning on Google Kubernetes Engine (GKE): Terraform for organizational scaffolding, async Python with Celery for orchestration, cloud provider SDKs (Google Cloud Platform / GCP, Amazon Web Services / AWS, Microsoft Azure), Postgres and Redis for state. Manages full lifecycle of ephemeral cloud accounts including provisioning, monitoring, recycling, and cleanup.",
        "Scaled to thousands of concurrent, session-based cloud environments per week. Designed the account pool and recycling architecture to optimize utilization and control spend across three providers.",
        "Architected provider-specific capacity strategies: OU-based account pools with recycling for AWS, resource-group provisioning for Azure, project-based allocation for GCP. Independently arrived at the same account pool pattern AWS uses internally for its own training environments.",
        "Built observability layer with Datadog: custom metrics, dashboards, alerting, and distributed tracing across the provisioning pipeline. Used tracing to find and fix bottlenecks, cutting environment startup time to under 30 seconds.",
        "Led Cloud Labs engineering: hiring, technical interviews, architecture decisions, cross-functional work with product and editorial to match capacity to demand.",
        "Managed IAM policies, access controls, and secrets rotation across all three clouds. Built abuse detection to track per-tenant resource consumption and flag anomalous usage.",
        "Conceived and shipped O'Reilly's first AI-powered Cloud Lab, an embedded coding agent (aider CLI + AWS Bedrock) that lets learners work with Large Language Models (LLMs) within seconds of launch. Built additional LLM labs on Azure OpenAI and GCP Vertex AI.",
        "Contributed resource handlers in Go to aws-nuke, an open source tool for programmatic cleanup of AWS resources across organizational accounts."
      ]
    },
    {
      "name": "Ciholas, Inc.",
      "position": "Software Engineer",
      "startDate": "2017-04",
      "endDate": "2018-12",
      "highlights": [
        "Built real-time front-end interfaces for an Ultra-Wideband (UWB) network monitoring system with millisecond-latency data visualization."
      ]
    },
    {
      "name": "Springleaf Financial Services",
      "position": "Programmer Analyst, Senior",
      "startDate": "2014-09",
      "endDate": "2017-04",
      "summary": "Promoted to Programmer Analyst, Senior in August 2015.",
      "highlights": [
        "Developed and deployed a hybrid mobile app for financial services in a regulated environment."
      ]
    }
  ],
  "education": [
    {
      "institution": "University of Southern Indiana",
      "area": "Computer Science",
      "studyType": "Bachelor of Science",
      "endDate": "2014"
    }
  ],
  "certificates": [
    {
      "name": "Google Cloud Associate Cloud Engineer",
      "issuer": "Google Cloud",
      "date": "2023-12",
      "url": "https://cloud.google.com/certification/cloud-engineer"
    }
  ],
  "skills": [
    {
      "name": "Languages",
      "keywords": ["Go", "Python", "TypeScript", "JavaScript", "SQL"]
    },
    {
      "name": "Cloud & Infrastructure",
      "keywords": [
        "Google Cloud Platform",
        "GCP",
        "Google Kubernetes Engine",
        "GKE",
        "Amazon Web Services",
        "AWS",
        "Microsoft Azure",
        "Kubernetes",
        "Docker",
        "Terraform",
        "multi-tenant SaaS",
        "capacity planning",
        "resource lifecycle management",
        "CI/CD",
        "Continuous Integration",
        "Continuous Deployment"
      ]
    },
    {
      "name": "Observability",
      "keywords": [
        "Datadog",
        "custom metrics",
        "dashboards",
        "alerting",
        "custom integrations",
        "distributed tracing",
        "logging"
      ]
    },
    {
      "name": "AI/ML",
      "keywords": [
        "Large Language Model integration",
        "LLM",
        "prompt engineering",
        "agent workflows",
        "Retrieval-Augmented Generation",
        "RAG",
        "LiteLLM",
        "AWS Bedrock",
        "GCP Vertex AI",
        "AI-powered applications"
      ]
    },
    {
      "name": "Backend",
      "keywords": [
        "REST APIs",
        "async Python",
        "Celery",
        "workflow orchestration",
        "FastAPI",
        "Django",
        "Node.js",
        "Postgres",
        "Redis"
      ]
    }
  ],
  "projects": [
    {
      "name": "aws-nuke",
      "description": "Open source contributor to aws-nuke, a cloud infrastructure automation tool written in Go",
      "url": "https://github.com/ekristen/aws-nuke",
      "keywords": ["Go", "cloud infrastructure automation", "open source", "Amazon Web Services"]
    }
  ]
}