Skip to content

Essays

Long-form writing on AI, philosophy, psychology, and systems thinking.

  • Remote Work Communication Infrastructure

    Redesigning communication infrastructure cut meetings and shortened decision latency, because decisions moved into written records. Remote work did not fail. The architecture was never built.

  • Observability as Epistemology for Distributed Systems

    Observability is epistemology for distributed systems. Logs, metrics, and traces construct knowledge about systems too complex for direct inspection.

  • Technical Debt Is a Loan, Not a Failure

    Technical debt is a tool, not a failure. Fowler's quadrant framework reveals four distinct patterns of debt, each demanding a different management strategy.

  • Platform Engineering Is Service to Others

    Platform engineering is duty made tangible. The best platforms reduce cognitive load so product teams ship business value instead of fighting infrastructure.

  • Camus, Sisyphus, and the CI/CD Pipeline

    The CI/CD pipeline is Sisyphus's hill. Meaning comes not from the permanence of your deployments but from the consciousness you bring to each push.

  • Conway’s Law Never Stopped Being True

    Conway's Law predicts that architecture mirrors team communication. Over half a century on, ignoring it still causes most architectural failures.

  • The Consulting Operations Paradox

    In my consulting work, the most valuable intervention was usually removing something, not adding it. Subtraction is harder to sell and easier to live with. Yet consulting economics reward building over removing.

  • The Spiritually Dispossessed Workforce

    The modern workforce is not merely exhausted but spiritually dispossessed, stripped of the connection between daily labor and existential purpose. The deepest workplace dysfunction is not structural but existential.

  • The Modular Monolith as Stoic Discipline

    Architectural simplicity is not a concession. The modular monolith demands more discipline than microservices and delivers more reliability for most teams.

  • Context Engineering Is the New Systems Design

    Context engineering treats what information reaches an LLM as an architecture problem. It reduced hallucination rates across enterprise deployments without changing the model.