Kant is usually thought of as too obscure and theoretical to provide practical advice, but I find a lot of practical advice in the Critique of Judgment on the question of how to be, or rather become, a genius. (I am working with the Bernard translation here). There are some preliminaries to keep in mind. First, Kant does not associate genius with high IQ. If you think that you have to have high IQ to be a genius then you are not talking about Kant's concept of genius. High IQ is overrated anyway. It is neither necessary nor sufficient for creative work. Before I go into his advice about becoming a genius I also want to make clear that Kant is just mistaken about scientists. He argues that a great scientist cannot be a genius. This is just a misconception about science. Science requires creative and original thinking every bit as much as art does. To be sure, creativity works differently in science than in the fine arts: for example you are not supposed to change your data set. There is perhaps less flexibility in science than in fine art. But, for now, let us just assume that everything Kant says about a genius in fine art could also apply to other practices including science, sports, business, religion, and philosophy itself.
Kant believes that genius is "a talent for producing that for which no definite rule can be given" and "is not a mere aptitude for what can be learned by a rule." This would eliminate people who are often seen as geniuses simply because they can memorize lots of information, follow certain rules, or play simple games like solitaire with great rapidity. None of this requires genius.
The mention of talent may seem to imply that genius must be something one is born with. However, Kant does not say directly that it is. He does speak of a pupil "endowed" with a "like proportion of mental powers" but I think this means that you need to find a teacher with "like proportion of mental powers" to yours: i.e. you need the right fit. Given that Kant believes that genius comes from the use of the faculties of imagination and understanding, and given that he believes that everyone has these faculties, I suggest that, on a Kantian view, anyone can be a genius, and everyone has the capacity (although usually undeveloped) to do original cutting-edge work. Granted, Kant says that genius is rare, but his system doesn't give us any reason to believe this is true.
How do you know that someone is a genius, then? Well, simply if that person is original in a way that is valuable: of genius, "originality must be its first property." Moreover, Kant clearly indicates that works of genius should not "spring from imitation" but should be "a standard or rule of judgment for others." We will see, however, that this later idea is qualified in an important way: the genius should not be a rule or standard for other geniuses if by this we mean something that can be copied or followed by way of clearly stated formulas.
Another factor is that the genius "cannot describe or indicate scientifically" how he or she brings about her products. Instead, the genius creates his or her own rules. I take this to mean that the genius creates his own definitions of key terms and concepts central to his own project and creates new ways of seeing and doing things. We shall see that this involves creating a world of his own, one based on the materials of the world we experience. An excellent example of someone who created his own rules was Van Gogh. In painting Starry Night, for example, he followed own rules and not the rules of accurate reproduction. He created an alternate world based on the world as he experienced it.
Also, like Plato before him in the Ion, Kant believes that the genius is inspired. That is, he speaks with approval of the origins of the word "genius." The word originally referred to a "guardian spirit given to the man at birth." This seems something like the Greek Muse, a source of inspiration, although personal. Again, I recommend that you not read this as saying that some people get a guardian spirit (or the physiological genetic equivalent) and some do not (although Kant might have meant this). It is quite possible that everyone is given such a guardian spirit at birth. In other words, in principle, everyone could act like a genius ... and hence could be a genius. This does not mean that everyone is capable of being a genius at every thing or at every time: there are certain things you are naturally fitted to do well, and certain other things not.
So here is the advice so far: Do not imitate. Create your own rules. Work towards originality. Produce models. Create your own world.
The next piece of advice is find a genius to be your teacher. Remember that although the genius artist creates his own rules, others can take that person's artworks as models. The genius artist prescribes his rule, a rule that cannot be reduced to a formula, but which is, rather, "abstracted from the fact, i.e. from the product." Others may then "try their own talent by using [this product] as a model, not to be copied but to be imitated." So the potential genius must find a genius teacher who should not be merely copied but imitated, not just imitated but "followed." Against this teacher he/she can test his or her own talent. As Kant says, "the ideas of the artist excite like ideas in his pupils if nature had endowed them with a like proportion of mental powers." If you want to be a genius you need to find your match in a great teacher, a great teacher for you.
Kant does not require that this be one teacher only. Nor does he insist that the teacher be living. The works of genius produced in the past can be models, can be the teachers. Indeed, he writes that "the models of beautiful art are the only means of handing down these ideas to posterity." A living teacher can teach you by means of his or her own works, but also through modeling his or her artistic practice, for example in the classroom.
On the other hand we cannot accept Kant's claim that "classical models are only to be had in the old dead languages." Clearly this is a prejudice of his time. However, we today are amazingly incapable of treating the dead languages (Latin, Greek, etc.) and writings in those languages as sources of inspiration, unlike the Renaissance and Kant's own time. We should not be so arrogant as to just assume that these should occupy the dustbin of history.
Kant offers another important piece of advice for being a genius. He observes that there is still a mechanical element in what you do; there are still skills to learn in your practice; there is still something "scholastic" to learn; you still need to follow certain definite rules in order to accomplish the purpose of your art. So you should not throw away "all the constraint of all rules," or to think that genius can replace "careful investigation by reason" where that is appropriate.
I want to add one more thing that Kant did not say but I think is consistent with the thrust of his approach. If you want to be a genius in Kant's sense find the hot spot of creativity in your city or neighborhood, and if there is nothing, move somewhere else where there is something. It might be a cafe where really interesting people bent on revolutionizing painting go to argue: that was the hot spot for the Impressionists in Paris in the late part of the 19th century. It might be a garage where people on the cutting edge in technology collect to pursue their dream. It might be a great start-up with a new idea. The hot spot is a place where you can find people who themselves can be models, contemporary models, in your quest for genius. Genius seems to thrive best in the hot house of interactive work and debate. What Kant teaches is that genius is not simply an individual trait but rather arises out a creative relationship between student and teacher, mentor and mentee, aspirant and model.
Finally, paragraph #49 constitutes the deepest thing Kant has to say about genius. Genius requires what Kant calls "spirit" and what is sometimes also translated as "soul." Kant calls it an "animating principle of the mind," which is to say that it brings things alive. What this does is put "the mental powers [imagination and understanding] purposively into swing, i.e. into such a play as maintains itself and strengthens the mental powers in their exercise." Moreover, this principle is "the faculty of presenting aesthetical ideas," i.e. of creating symbols (and works of art can be examples of such symbols), imaginative products which send the mind into much thought, none of it definite or completely intelligible by language.
It is here that Kant tells us that the imagination, in this productive use, "is very powerful in creating another nature" out of the material of actual nature. It remolds experience using metaphors and analogies. Such symbols "strive after something which lies beyond the bounds of experience" and try to get as close as possible to present what Kant calls the "concepts of reason," by which he means transcendent things such as God, soul and immortality, as well as all the most fundamental ideas by which and under which we live; for example, death, envy, love, and fame. In doing this, the imagination "brings the faculty of intellectual ideas (the reason) into movement": i.e. it enlivens these ideas. In sum the genius is someone who creates things that connect us in this way to what is transcendent but ultimately inexpressible.