<p>
As a caveat, opinions vary considerably as to what constitutes “an amazing graduate school”… but I am going to assume that you mean “most selective” because that is at least an objective target.</p>
<p>Most grad programs look at applications in two rounds. The first round is a down-select, where they use the simple numbers (GPA and GRE, generally) to pare down a huge stack of applications to a manageable number - they might take the top X%, or you might need a Y+ GRE and a Z+ GPA, it varies. The second round is an up-select where those with spots to fill look at your area of interest, preparatory coursework, research effort, and (most importantly) letters of recommendation. So really, to get into a “most selective” grad program, you need to be pretty darned strong on all of these.</p>
<p>
I encourage you to shoot for a 4.0 in the same way that I would encourage a baseball pitcher to shoot for a perfect game - it is a laudable goal, but don’t stress too much if you fall short. Really, anything above a 3.5 will get consideration just about anywhere, and anything above a 3.8 generally comes down to classes that your grad program won’t care about. If you get a C in Anthropology 146, no EE grad program is going to notice. But if you are going for antenna design with a B+ in RF and Microwave Circuitry then you might have a problem.</p>
<p>As to how to do it? Pretty simple, really: put academics first. Study. Show up for every class. Take notes. Engage the professor and TA. Do the homework early. Go to office hours. Ask for help when you need it, and do so intelligently. Study for exams well ahead of time. When you make mistakes, don’t repeat them. And focus on learning the material, not just passing the class - because this stuff is going to come up again in the future, and you really want to show the professors that you learned, not that you can beat the system.</p>
<p>None of this is rocket science, it is just that most people will not do it. It takes discipline, not a strong suit amongst the 18-25 set.</p>
<p>
Internships are all but useless for grad school*. What you want is research. You want to get into a professors research group as early as you can - sophomore is great, junior is more realistic, senior if nothing else. Do some work over the year, and more over the summers if possible. There is something called the REU program, funded by the NSF - check it out. But research is king, internships teach you (generally) how to work in industry.</p>
<p>*: There are a small number of internships that are really research opportunities IN industry, but they are the exception, not the rule.</p>
<p>
Yes - don’t get too stressed. That up-select process means that admissions is extremely individualistic. There is no one path, one method, one course. Most things you do will be disliked (or ignored) by some professors and praised by others. What you really need is a few good numbers (GPA and GRE) and a few respected researchers (read: professors) who will personally attest to how you are an exceptionally hard worker with talent and passion in the area of … whatever. All the rest is frippery.</p>
<p>
Mostly that you know what you are getting into. That you understand your position and what you are asking to be a part of. And that you can handle it. This isn’t undergrad, they are not looking for stunning prose and lofty dreams, they are looking for a pro, someone who will come in, do the work, advance the lab (and their advisor), and go on to a glorious research career.</p>
<p>But don’t kid yourself - the letters of recommendation will carry ten times the weight of anything you can write about yourself.</p>
<p>
You’re welcome.</p>