Posts Michael Orlando

Colophon — mikeaorlando.com

This site is not a static artifact. It is the output of an evolving system designed to make ideas observable, structured, and useful to other people. The system exists to support collaboration, maintain accuracy, and keep the content improvable over time.

Purpose

The system exists to ensure that:

All pages of mikeaorlando.com and associated media are publicly available, human-audited, accurate, and as useful as possible—allowing readers to quickly find relevant information related to Michael Orlando and his work.

This is achieved by separating source truth, transformation, and publication, while enabling both humans and AI to participate in the process.

Core Principles

  • Source-first editing — Content is never authored directly on the website.
  • Human accountability — All published content is reviewed by a team.
  • System integration — Content lives within broader operational systems.
  • AI-assisted, not AI-controlled — AI proposes; humans decide.
  • Portability and ownership — Content can be restructured or republished at any time.
  • Repeatability — The process can be reused across brands and teams.

System Overview

The system consists of four layers:

  1. Source of Truth
  2. AI Transformation Layer
  3. Version Control & Review
  4. Static Publishing Infrastructure

1) Source of Truth

All content originates in operational systems, not in the website itself.

Authoritative systems:

  • Confluence — Structured written content, brand guidelines, messaging
  • Google Drive — Media (images, video, documents)
  • Jira — Work tracking and coordination

Content created here is considered:

  • Accurate
  • Representative of reality
  • Ready for downstream use

Examples of source content:

  • Notes from real-world events
  • Meeting summaries
  • Scripts
  • Media assets
  • Concept documentation

2) AI Transformation Layer

AI tools are used to interpret, organize, and propose how content should appear publicly.

Tools used:

  • OpenAI Codex
  • Claude

Inputs:

  • Source content (via MCP integrations such as Atlassian Rovo and Notion)
  • Brand and marketing context from Confluence
  • Existing website structure

AI responsibilities:

  • Propose new pages
  • Suggest edits to existing pages
  • Recommend structure and navigation
  • Generate drafts and content layouts
  • Map raw content to audience-specific messaging

AI constraints:

  • Cannot publish directly
  • Cannot override brand or structure rules
  • Must operate within provided context

3) Human Review & Version Control

All AI-generated or modified content is reviewed before publication.

Responsible team:

  • AspirationalX MarCom (Marketing & Communications)

Responsibilities:

  • Validate accuracy
  • Ensure alignment with brand voice
  • Confirm usefulness for the reader
  • Approve or reject structural changes

Version control:

  • GitHub repository
  • Changes committed to a staging branch
  • Reviewed before merge to main

4) Publishing Infrastructure

The site is built using a static architecture optimized for performance, reliability, and control.

Current stack:

  • Hugo — Static site generator
  • AWS S3 — Content hosting
  • CloudFront — Global CDN and caching
  • Route53 — DNS management

Deployment process:

  1. Content changes committed to staging
  2. Hugo builds static site
  3. Deployment to staging environment
  4. Validation of functionality
  5. Merge to main branch
  6. Production deployment

Content Flow (End-to-End)

  1. An event, idea, or insight occurs

  2. Content is captured in:

    • Confluence (text)
    • Google Drive (media)

  3. A human prompts AI with a specific request

  4. AI proposes:

    • New pages or edits
    • Structure and placement

  5. MarCom team reviews proposal

  6. Approved changes are:

    • Converted into Hugo-compatible markdown
    • Committed to GitHub (staging)

  7. Site is built and deployed to staging

  8. Changes are verified

  9. Merged to main and deployed to production

Site Structure & Governance

The structure of the website is not arbitrary.

It is defined by:

  • Audience definitions
  • Messaging frameworks
  • Campaign strategies

These are documented in Confluence and used as context for both:

  • Human decision-making
  • AI-generated proposals

This ensures:

  • Consistency
  • Alignment with brand intent
  • Avoidance of fragmented or redundant content

Voice and Style

Each brand, including the Michael Orlando brand, has a defined:

  • Voice
  • Tone
  • Messaging strategy

These are maintained in Confluence and enforced through:

  • AI prompting context
  • Human review prior to publishing

Evolution of the System

The current system is the result of several iterations:

Phase 1 — WordPress (Bluehost)

  • Direct editing in WordPress
  • Simple, but limited flexibility and structure

Phase 2 — WordPress (AWS Lightsail)

  • Improved infrastructure control
  • Still constrained by CMS limitations

Phase 3 — Notion + Sotion

  • Content-first workflow
  • Better collaboration
  • Limited control over final output

Phase 4 — Hugo + AWS (Current)

  • Full control over structure and deployment
  • Separation of content and presentation
  • Integration with AI and operational systems

Why This Architecture Exists

This system solves several problems:

  • Prevents content fragmentation across tools
  • Enables teams to collaborate without publishing prematurely
  • Allows AI to accelerate content creation without risking accuracy
  • Ensures all published content is intentional and reviewed
  • Provides a scalable model for multiple brands and sites

Tools Used

Content & Collaboration

  • Confluence — structured documentation
  • Jira — workflow and task management
  • Google Drive — media storage
  • Apple Notes — quick capture (intermediate)

AI & Transformation

  • OpenAI Codex
  • Claude
  • MCP integrations (Atlassian Rovo, Notion)

Development

  • Visual Studio Code
  • GitHub

Publishing

  • Hugo
  • AWS S3
  • CloudFront
  • Route53

Previous Systems

  • WordPress (Bluehost, Lightsail)
  • Notion + Sotion
  • Buffer (distribution)

What This Enables

  • A team can maintain a coherent brand over time
  • Content can be created continuously without immediate publication
  • Knowledge is preserved in structured systems
  • Websites become outputs of a broader operational model
  • The system can be replicated for other brands and initiatives

Reproducibility

This system can be recreated by:

  1. Establishing authoritative content systems (Confluence, Drive)
  2. Defining brand and messaging frameworks
  3. Enabling AI access to structured content
  4. Using AI to propose structured outputs
  5. Enforcing human review
  6. Publishing via a static site generator and cloud infrastructure

This colophon reflects the system as it exists today. It is expected to evolve as tools, workflows, and understanding improve.