As cool as it sound, do we actually need it? also what problem are we solving?

Performance is probably the only angle that has some potential but only short requests, the db and other external IO/services is a bigger bottleneck.

Because the performance of existing web frameworks is... _absymal_. They are nowhere close to how fast they can be. And I don't mean slower by a bit, I mean close to a factor of 10x. My in-house framework can serve 600-700k requests/s on a single core, while beast (the best c++ framework i could identify) does 250k at best. nginx/apache and other prebaked servers are even slower, at 100k req/s. Tells you all you need to know.