← All posts
Hello, World: Why I'm Starting This Blog

Hello, World: Why I'm Starting This Blog

Welcome. This blog started as a notebook I was going to keep for myself, and then I thought: why not let other people read it too?

I’ve spent the last several years working in application and product security: threat modeling, code review, mobile security, and looking for scalable, high-impact issues. But the thing I’ve learned most is that the most useful knowledge isn’t the polished version. It’s the messy part: what I tried first, what didn’t work, and what finally did.

Why I’m Writing This

I keep forgetting things.

I’ll spend a whole afternoon figuring out why a setting doesn’t apply, or why a network share says “no permission” when the password is correct. I fix it, feel good for about ten minutes, and then three months later I hit the same problem and have no idea how I solved it.

So this blog is a digital diary. It has three jobs:

  • Remember. Writing it down is how I make sure the lesson sticks.
  • Keep the evidence. Failures are as important as successes. A record of a wrong turn is often more useful than the right answer, because the next person probably will take the same wrong turn.
  • Share it. If something I learned saves someone a few hours, the diary has done more than I expected.

Nothing here is too small to write about. A security finding, a new tool, a home network that refuses to share a folder: they all count. If I learned something while building it, it belongs here.

Where It Starts

The first real entry is a small project, not a security write-up. I wanted a shared folder on my Windows PC that every device in the house could reach, without buying a NAS. The post covers the setup and, more honestly, the three or four problems I hit along the way:

It’s a good example of what this blog is for. The setup itself took minutes. The debugging took much longer, and that’s the part I most want to remember.

What to Expect

The themes I’ll be writing about:

  • Threat modeling in practice. Not the textbook version, but what actually works when you’re sitting with an engineering team.
  • Mobile and web security. Real vulnerability patterns, and how to fix them centrally instead of one ticket at a time.
  • AppSec at scale. Using automation and AI-driven tooling to keep security up with engineering speed.
  • Home and personal tech projects. Networking, homelab work, and other small experiments, including the ones that fail.

Every post follows the same approach: what I was trying to do, what went wrong, what I checked, and what I learned.

How This Is Built

This site is a hand-built static page, and the blog runs on Jekyll, which GitHub Pages builds automatically. Writing a new post is just adding a Markdown file like this one. No build server, no database.

A diary is only useful if it’s honest about the failures.

More soon.