Great question! There are a number of candidate solutions to the problem of finding relevant decks. And as mentioned by @O3nnM7LP5GzFMBS, the AnkiHub team has implemented a solution that, if it gained wide enough usage and they work out the bugs, could be a substantial improvement over AnkiWeb. Unfortunately, I see a difficult road ahead of them convincing the entire community to jump ship, mainly because their system costs money, and the existing system is free. People say a new product needs to be 10x better than the existing alternatives to get widespread adoption, and I think that adage makes the implicit assumption that both the old thing and the new thing cost about the same.
AnkiWeb can certainly be improved, although it is fantastic in that it is free and searchable at all, and that there are a large number of creators who live there. The search facility is extremely rudimentary in modern terms, and seems almost purposefully limited. There is also no good way to browse decks before you download them, or version your releases if you’re a maintainer. It’s also not exactly ideal that all of that data is at the mercy of whoever runs the AnkiWeb servers (is it just Damien? I don’t know).
My hope is that eventually, the deck maintainer community makes a soft, gradual transition to a platform like GitHub, GitLab, or sourcehut. These platforms have extremely mature, sophisticated and reliable search mechanisms, and they have browsing functionality built-in. They also have the advantage of supporting collaboration with git, which is an extremely robust, powerful distributed version control program. These services are free, have interoperable decentralized alternatives, and really care about preventing data loss. They also make it very easy to build a community around a deck, solicit user contributions, correct errata, etc.