I was clearing out the garage and found a dusty box of old DVDs. Not rare, not collectible, not even particularly sentimental. Just the usual mid-2000s assortment of things you once owned on plastic because Netflix didn't exist yet. The sensible thing to do would’ve been to stick them on eBay as a job lot, or charity shop the lot and move on with life. Obviously, I did neither.
Instead, I plugged in a barcode scanner that I bought on eBay in 2019 and has been sitting in my office cupboard ever since.
I’ve always had a soft spot for turning small, boring problems into fully engineered systems. It’s not about efficiency in a strict time-saved sense, it’s about the satisfaction of flow. Scan thing, data appears, decisions happen automatically, no thinking required. So rather than manually typing DVD titles into eBay one by one like a normal person, I decided this deserved a pipeline.
The barcode scanner kicks things off. Each scan hits a small script that looks up the DVD via an external API, pulls the title, year, cover art, and basic metadata, and drops it into a local inventory table. From there I can see what I’ve got, spot duplicates, and decide what’s worth listing versus what’s destined for the charity pile. No spreadsheets. No copy-paste. Just scan, beep, next.
A draft eBay listing is generated automatically with a sensible title, category, condition, and pre-filled description. I still press publish, because human in the loop and all that, but by that point the work is basically done.
Is this wildly excessive for a box of old DVDs? Obviously. Did it take longer to build than it would’ve taken to list them manually? Almost certainly. Will I ever need this exact system again? Probably not. But that’s not really the point.
I like building systems. I like small automations that remove friction. I like the feeling that once something exists, it keeps working quietly in the background, whether it’s powering a SaaS product or helping me clear out a garage. There’s something deeply satisfying about turning even the most mundane tasks into a clean, repeatable flow.
Most people see a box of DVDs and see clutter. I see an excuse to wire together hardware, APIs, a bit of glue code, and a tiny workflow that didn’t exist yesterday. It’s the same instinct that makes me build internal tools, automate admin, or tinker with electronics on a Sunday afternoon. The scale changes, but the motivation behind it is the same.
What’s changed recently is how easy it is to go from idea to working system. Before AI, I could absolutely have built this. It just would’ve taken the best part of a day rather than half an hour, and my attention would almost certainly have wandered off long before it was finished.
AI is incredibly good at lowering the activation energy for these kinds of ideas. Things that used to feel mildly interesting but not quite worth the effort suddenly become trivial to try. Slightly silly ideas turn into real, working systems with very little friction.
If you do that often enough, one of them might turn out to be genuinely useful. And even if it doesn’t, you still end up with a working system and a tidier garage, which feels like a decent trade.
And if you are in the market for some DVDs, please check out my eBay page 😄
Ross