I looked at a few companies in this space when I was at a VC, and the fundamental problem was that they didn't provide enough value to the end developer.

Aggregator APIs tend to provide extremely reduced functionality compared to the source API, and the ease-of-use doesn't compensate for this. In most cases it just makes more sense for the developer to spend a day building a custom adaptor for the API.

That said I think the SQL front-end is an interesting twist. While developers find it easier to use the source API there are many people (i.e. business analysts, etc.) who can't program but can use SQL and that might be an interesting market to go after (and also a market more willing to pay).