Skip to content
Back to overview
Projects

Selected engineering work

Six case studies covering platform tooling, infrastructure as code, cost engineering, and putting AI agents into workflows without giving up control of them.

  1. AI in engineering workflowsRead ↗

    AI Pull-Request Reviewer

    An agent that reviews every pull request against its Jira acceptance criteria, and says Unknown when it cannot tell.

    70+
    repositories, up from 30
    Claude CodeGitHub ActionsJira APIGitHub API
  2. AI in infrastructure operationsRead ↗

    Namespace-Scoped MCP Servers

    Letting AI agents operate Kubernetes without letting them operate it unsupervised.

    RBAC
    on every mutating path
    MCPKubernetesRBACEKS
  3. Platform & developer experienceRead ↗

    Backstage Golden Paths

    Cutting new-service provisioning from twelve days to under one, by making the safe path the fast one.

    12d → <1d
    to provision a new service
    BackstageTerraformAWSGitOps
  4. Cost & capacity engineeringRead ↗

    Karpenter and Spot Capacity

    Cutting Kubernetes compute spend by 25-35% without moving critical services off On-Demand.

    25-35%
    Kubernetes compute saved
    KarpenterEKSEC2 SpotKubernetes
  5. Infrastructure as CodeRead ↗

    CloudFormation to Terraform

    Rebuilding a live EKS environment as 50+ Terraform-managed resources with zero disruption.

    Zero
    disruption during migration
    TerraformEKSIPv6OIDC
  6. Platform engineeringRead ↗

    This Site

    A personal site built as a platform exercise: one codebase, two deployment targets, and an AI twin that cannot be forged.

    2
    deploy targets, one codebase
    Next.jsDockerTerraformS3