Enabling AI Capabilities in Docs as Code Projects
Nature of Docs as Code Projects
Docs as Code is increasing in popularity within the marketplace and has been adopted as a documentation strategy by some of the largest enterprises as well as by SMBs. The main reasons for this include:
• the need to align software development with the documentation process
• need to have software developers and engineers working in simple markdown content authoring
• adopted where CI/CD delivery is required (primarily Tech SaaS but also other sectors)
• cost-effective as tools used currently for code development are adopted — GitHub, GitLab, BitBucket, etc.
Docs as Code is a custom development approach and requires a technical team to develop and support the environment.
The increasing importance of AI within documentation projects
Enterprises today are very motivated to deliver AI capabilities in documentation to deliver:
• Chatbots
• Dynamic Support sites
• In Context Help
• Enhanced search
The increasing importance of deploying AI projects within documentation projects has led to issues with this approach. Markdown, by its very nature, creates static file-based content from which AI can extract intelligence in limited ways.
Enterprise support content has issues with content references based on how the software operates while containing unique definitions and relationships between the content that AI cannot recognize within the content itself.
For AI interpretation, metadata is the key to understanding the relationships between content and the context. The C Suite demands that these strategies be deployed as there is a competitive advantage in delivering these capabilities. AI in documentation projects brings substantial cost reductions in content management and distribution.
How to deliver AI projects in Docs as Code Environments?
The key is to deliver enhanced Metadata content while still allowing engineering teams to generate content in Markdown. This allows for intent-based content using specific tags to be created and the ability to wrap intuitive content that can be indexed by the AI engine.
Metapaercpet metR is a self-service modular content migration solution that migrates Markdown, HTML5, and Word/Google Docs content into a structured modular format.
metR also provides complementary tools for managing, publishing, editing, and versioning these newly structured documents in your organization’s GitHub.
Advantages of this approach
• It allows content to be generated as usual while the engineering team continues working on the existing Markdown content.
• metR utilizes existing tools and open-source solutions (including GitHub as the content repository) — a very cost-effective, rapid-to-deploy approach.
• metR is easily adopted by internal teams supporting your current environment. It is a cost-effective ongoing approach to enhancing the current support documentation.
• training and support for internal resources are needed to ensure the rapid and effective adoption of this approach.
Call Metapercept to discuss your requirements; we can set up a POC to show the approach in action.