How I Prepare These Posts | DineshStackSkip to content

How I Prepare These Posts

There's a lot of thin, AI-generated tutorial content online right now that reads fine and doesn't hold up. I'd rather tell you exactly how this blog actually gets made than let you wonder.

Last updated: July 2026

Where a topic comes from

Most posts start from something I actually built — a ride-hailing platform, an e-commerce checkout, an AI voice assistant, a production server I manage day to day. When a post uses real code from client or production work, I rewrite it to remove anything identifying (names, brands, internal URLs) but keep the actual logic and structure intact. The goal is code you could genuinely learn from, not a sanitized success story with the interesting parts removed.

Checked against the source, not memory

Frameworks move fast, and I'd rather say nothing than say something outdated. Before publishing anything about a specific framework version or a recently shipped feature, I check it against the current official documentation and changelog — not what I remember from a previous release, and not what a first-draft summary claims. If I can't verify a claim, it doesn't go in.

Where AI actually fits in

I use Claude and other AI tools while researching and drafting — the same way I'd use any tool that makes the work faster. I'm not going to pretend otherwise, and I don't think I should have to. What I won't do is publish something an AI produced wholesale and never checked. Every example here either comes from something I've actually built, or was verified against a primary source before it went live. The tool doing the typing isn't the part that matters — whether the content is true is.

What this looks like in practice

Rather than just assert all of this, here's a real example: a tutorial that was rewritten from unfilled placeholder text into a real deep dive built around anonymized jobs from a production ride-hailing platform, with every framework-specific claim checked against the current Laravel docs before publishing.

Read the Laravel Queues Deep Dive

If a tutorial saves you real time

Some tutorials have a companion code pack on Ko-fi — the full working project, if rebuilding it yourself isn't what you're after right now. That's the main way to support this blog directly at the moment. As traffic grows, I may open up other ways to support the time (and the paid tools) that go into these — but that's a decision for later, not something I'm asking for today.