Statement on AI Usage

I started working on ConnectHV long before generative AI first hit the scene. The first several versions of this web platform were developed without any AI augmentation, for the simple reason that it didn't exist.

This latest version of the project (circa early 2026) has been developed with the assistance of several AI models, including:

• ChatGPT (primarily 5, 5.1, and 5.2)
• Claude (Sonnet 4.5 and Opus 4.5, via Claude Code)
• Cursor Composer 1
• Google Gemini (via Google Antigravity)

What I use AI for

I've used these tools to assist with programming tasks, including:

• Fixing small bugs
• Writing new features
• Refactoring messy code
• Red-teaming security vulnerabilities
• Improving test coverage

Crucially, these are all things I'm already skilled at, so I'm able to use AI as an accelerant and "intern" rather than a replacement for net-new things I couldn't possibly evaluate the quality of.

What I haven't used AI for

I haven't used AI to create any graphics or perform any tasks that I don't know how to perform manually, and I haven't used it for any tasks that I don't know how to evaluate the output of.

A software engineer is responsible for the output they produce, regardless of how it's produced (manually or with AI augmentation). Therefore, I only use AI for tasks of which I can safely evaluate the output.