Why this blog

I spend most of my time building two AI-related projects. I want a place to write about the decisions, dead ends, and small wins that happen in between commits. This blog will cover:

• Project updates — launch notes, feature retrospectives, and things that did not work.
• Technical write-ups — architecture choices, weird bugs, and lessons from building in the open.
• Thinking out loud — ideas about AI interfaces, prompt engineering, and reading as a learning tool.

What I am building

Two things right now:

1. promptforge

A prompt-generation tool for landing-page builders on AI front-end platforms. It focuses on information architecture and conversion — not just pretty text, but pages that actually guide users to act. If you care about how AI-generated sites can still feel intentional and human, this is the space I am exploring.

2. readingcoach

An interactive AI companion for non-fiction books. It uses a five-layer architecture and an adaptive depth-reading model to help readers engage with dense material without getting overwhelmed. I am currently in the PRD phase, having gone through two rounds of deep iteration with challenger reviews and defense responses. The design is getting sharper, and I will share milestones here as it moves from paper to prototype.

What to expect

No fixed schedule. I write when something is worth capturing. Posts will be short and practical. If you are building AI tools, reading hard books, or just curious about how indie projects get made, you might find something useful here.

You can follow along via the RSS feed or check the project tags for updates on specific tools.

Thanks for stopping by. More soon.