I Got Tired of Formatting Blog Posts, So I Built My Own Publishing Tool with AI

One article. Six design styles. A reusable publishing workflow.

My Problem Wasn't Writing. It Was Everything After Writing.

I enjoy writing about technology, AI, education, and my experiments with digital solutions.

But every time I wanted to publish an article on Blogger, the same repetitive work appeared.

Formatting headings. Adjusting layouts. Adding branding. Inserting my standard header and footer. Making everything look consistent.

I thought:

Why am I repeating the same work when a computer could do it for me?

So I decided to build a little tool.

Not a massive application. Not another expensive subscription.

Just something useful.

Meet My Little Publishing Factory 🏭

I developed a Python-based article converter as part of my self-learning journey, using AI as a learning and development partner.

Here's my workflow:

1. Write my article in Microsoft Word or Markdown.
2. Run a simple Windows batch file.
3. Choose one of six predefined design styles, or let the converter select automatically.
4. Generate Blogger-ready HTML in a text file.
5. Copy, preview, and publish it in Blogger.

That's it!

My converter already knows my preferred layouts, branding, headers, and footers.

The six styles cover different kinds of content, from professional business articles and educational posts to technical tutorials and editorial writing.

For example, I recently selected my Navy Orange Editorial design.

The converter processed my Word document and produced the formatted HTML.

No need to rebuild the entire design for every article.

And importantly, I still review the output before publishing.

Here's What I Learned

I'm not a traditional software developer.

I've explored Python, databases, cloud platforms, automation, and web technologies through self-learning and hands-on experimentation.

AI has helped me explore ideas, understand problems, and develop practical solutions.

But I discovered something more valuable than simply asking AI to write code.

The real skill is learning to explain what you want to accomplish.

Instead of saying:

"Build me something with Python."

Try explaining:

"I publish articles regularly. My formatting is repetitive. I want a tool that takes a Word document, applies my preferred design, and generates HTML that I can paste into Blogger."

Now you have a real problem, a clear workflow, and an outcome you can test.

That's where useful building begins.

You Can Try This Approach Too

You don't have to start with a complicated application.

Find one task you repeat regularly.

Document how you do it today.

Ask AI to help you simplify the process. Test the result, understand what it does, improve it, and keep yourself responsible for the final output.

Your first tool might save only five minutes.

But the next time you use it, those five minutes are saved again.

The goal isn't to automate everything. It's to stop repeating work that no longer needs your attention.

My Bigger Lesson

I used to think technology was mainly about learning tools.

Now I increasingly think it's about understanding problems.

You don't need to be an expert to begin experimenting.

You need curiosity, clear thinking, patience, and the willingness to learn when something doesn't work.

I'm still learning. Still making mistakes. Still building.

And yes, sometimes a simple script gives me more satisfaction than an entire complicated application. 😄

You can explore my published articles at DeTLeng Insights and my broader digital learning projects at DeTLeng.

What repetitive task would you automate if you had your own little AI-assisted development partner?

I'd love to hear your ideas in the comments.

Keep learning. Keep experimenting. Keep building.

— A self-learner turning everyday problems into practical digital solutions.