If you want to splice one gene that is feasible but there is only so much you can accomplish. If you were going to make somebody much more intelligent, stronger, taller, whatever you will have to splice a large number of genes and there is always the possibility’s you screw it up and the child is worse rather than better, maybe catastrophically.
What is feasible now is you could use IVR to make a large number of eggs, sample the DNA of the eggs, score the embryos based on what traits you think are desirable, and pick the best. That seems pretty safe but it’s arguable how big of a difference it could make.
An even more exotic possibility is this: it is possible to harvest gametes from an unborn fetus so one could do that selection, and then do the same selection again, applying several generations worth of selection to produce a very high scoring embryo.
Ultimately, the issue is one of "designer babies" is one of marketing.
For just a few thousand dollars, other mammals, mostly pet dogs, can be cloned. Scientists have been cloning rats and mice for decades, really. There's absolutely no technical reason to not clone humans, and we know that other human fertility techniques get abused. How many stories have you seen about some creepy fertility doctor using his own sperm to impregnate hundreds of women?
Human cloning's been done. So has CRISPR editing.