About MergeJPG

In September 2024, I was working on a side project - helping a friend set up her Etsy store. She needed to merge product photos to show color variations, and I told her to use one of those online tools. She came back confused: "Why does it need to upload my photos? And why is there a watermark?" I didn't have a good answer.

That night, I got curious about whether browsers could handle image processing locally. Turns out, the Canvas API has been around for years and is pretty powerful. I spent that weekend building the first version of MergeJPG - mostly learning by breaking things and reading Stack Overflow. The first version was buggy (images sometimes merged upside down, which was embarrassing), but it worked.

I soft-launched it on Product Hunt in October 2024. The first week was quiet - maybe 30 people tried it. Then someone shared it on a Reddit thread about privacy tools, and traffic spiked. People seemed to really care about not uploading their files. As of December 2024, the site has processed over 50,000 image merges, all happening entirely in users' browsers.

The more I looked at existing image tools, the more problems I noticed:

• Most tools upload your files to their servers. For memes, fine. For client work or personal photos? That felt wrong.
• The free ones add watermarks or have tiny file limits. The paid ones want $9.99/month for a feature that should be simple.
• Tools like Photoshop are overkill for just merging two images. I love Photoshop, but it's not worth opening for a 30-second task.
• The UIs were confusing. Too many options, too many steps. I just wanted to drag, drop, merge, download. That's it.

I made some specific technical decisions to solve these problems:

It's mostly just me, Alexander Duan. I'm a full-stack developer who's built a few SaaS products before. I work on this in my spare time, usually late at night when I probably should be sleeping. My day job is building web applications, so this scratches the same itch but with a simpler, more focused scope.

I occasionally get help from designer friends when I mess up the UI (which happens more than I'd like to admit). But the code, server management, and decision-making is all me. That means it moves slower than a VC-funded startup, but it also means I can keep it simple and free without worrying about investor returns.

My goals are pretty straightforward:

• Prove that simple image tools don't need to be complicated or privacy-invasive. Client-side processing works.
• Keep it free and accessible. I hate when basic tools are locked behind subscriptions. This won't be one of those.
• Build something that solves one problem really well, instead of trying to be everything to everyone.

I'm working on PDF export and better grid customization. But honestly, I'm trying to resist feature creep. Every time I think "this tool needs X," I remind myself why I built it in the first place - to be simple. So I'm slow to add features, which frustrates some users but keeps the tool focused. If you need advanced features, tools like Photoshop or GIMP exist for a reason.