Great job! A categorization system (especially if it can be crowdsourced/edited via github) would be great, since these datasets are likely to be useful across

Great job! A categorization system (especially if it can be crowdsourced/edited via github) would be great, since these datasets are likely to be useful across multiple domains - and not just "CV" or "NLP" - I'm thinking "Stocks", "Finance", etc. Also https://registry.opendata.aws/ https://registry.opendata.aws/ from AWS has a lot of datasets that could either be included en masse, or even linked to for the page to be more comprehensive. I like their categorization system (tags/labels) as well. They also have usage examples which is excellent to get a sense of what the data is/useful for.