tzoSkatzo

Thanks for your feedback!

Actually I've thought of that as well, but my main concern is that it's a bit difficult to find out median prices for "non-standardized" items, and I want to offer a price comparison that is meaningful for users. So cellphones are possible, but prebuilts could be trickier. It would be an interesting challenge to tackle eventually, if eBay offers me more API calls than they currently offer.

Yeah, it's funny the eBay API tends to return mostly irrelevant listings, and I have to classify them. Even then they are like 95% accurate, one thing I could potentially improve is use a classifier per model instead of per part type.

However, your feedback is noted, I'll try to add this asap.

Thanks for the feedback, I agree so I updated the title.

Thanks for your question.

That was a real concern of mine, for a long time. Business Wise: 1. eBay has a really good buyers protection. I've had items that were not delivered or dysfunctioning, and I got a full refund. No questions asked. 2. Make sure to check the sellers rating. - pcprice.watch also includes the sellers' rating.

But I understand that you are more asking about my approach for a GPU to work hardware-wise? Technical testing for GPUs: 1. I do PCB inspection, and I check the power connectors PCIe contracts. 2. I found out GPU-Z to be useful 3. And maybe the most telling gaming benchmark suite: 3DMark or gaming sessions

So far I haven't had any incidents that had made me lose money, and I do flip PC's quite often :)

Technical detail for those interested:

Architecture: - Frontend: Vue.js with Tailwind CSS, hosted on Vercel - Backend: Python scripts running on Oracle Cloud VM - Data refresh: Every 8 hours via cronjob

The most interesting challenge was filtering legitimate PC part listings. eBay search isn't perfect - searching for "RTX 4090" returns lots of irrelevant items like brackets, broken cards, or water cooling kits.

I solved this using custom ML classifiers for each component type (GPU/CPU/RAM/Motherboard). Each classifier is trained on labeled eBay listings to identify legitimate parts.

This automated filtering is crucial since the scanner processes thousands of listings daily across different eBay markets and languages.

One other issue, is bypassing the eBay API limitations, since I query for thousands of items I need to somehow optimize it. So far, I'm caching useful information like Shipping, but eventually if I want to add more items and/or marketplaces, I will need to either implement a key rotation or ask ebay to extend their API limits.

Currently, I scan for around 10k eBay listings along with their shipping info.