Documentation Methodology
At AWS, I practiced a docs-as-code workflow, treating documentation with the same rigor as software engineering. My documentation process involved:
- Source control: All documentation maintained in GitFarms (AWS internal Git), with pull request reviews, version history, and branch-based development
- Structured authoring: Oxygen XML Editor for DITA-based content, enabling content reuse, conditional processing, and consistent information architecture
- Code validation: Deep-diving into source code, testing examples, and working directly with engineering teams to validate every procedure and code sample
- Continuous delivery: Documentation shipped alongside feature releases, with automated builds and publishing pipelines
Tools & Platforms
Documentation Stack
Authoring: Oxygen XML Editor, DITA, Markdown
Source control: GitFarms (Git), code review workflows
AI tools: Custom-built documentation pipelines using Claude, MCP, and Python
(see AI-Powered Tools)
Analytics: SQL and Python for documentation health metrics analysis
Collaboration: Quip, Wiki, cross-team review processes
AI-Powered Tools
Beyond documentation, I designed and built AI-powered tools to automate complex workflows. These tools handled drafting, ticket resolution, and editorial review, achieving measurable time savings and consistency improvements.
Featured Projects
The following projects demonstrate different aspects of technical communication at AWS, from product how-to guides to architecture documentation and cross-functional process work:
BYOI: Custom Container Environments
End-to-end documentation for bringing custom Docker images into SageMaker AI Studio. Authored complete how-to guides, worked directly with engineering to validate examples.
View sample →Remote Access: Local VS Code to Cloud Spaces
Documentation for the #1 requested SageMaker feature—connecting local VS Code to cloud-based Studio spaces. Complex security architecture, multiple connection methods.
View sample →Cluster Management: Task Governance Architecture
Complex system documentation for SageMaker HyperPod task governance. Enterprise-scale cluster management made understandable for administrators.
View sample →Trusted Identity Propagation: Enterprise Identity Management
Enterprise identity management across AWS services—enabling fine-grained access control, user-level auditing, and seamless integration with S3, EMR, Redshift, and Lake Formation.
View sample →Onboarding Docs: SageMaker and AWS-Wide Improvements
Cross-functional documentation improvements that impacted AWS-wide. Identified friction points, proposed solutions, drove org-wide changes.
View sample →