Advice to Incoming Graduate Students

Today, a friend of the family reached out for advice on coming to Georgia Tech for the ECE PhD program. He’s married with a young child.

This post does not question the premise that graduate school is a good thing to do. I have other opinions on that which I’ll reserve for a later date.

Here’s my advice to incoming graduate students:

There is a large range of PhD program lengths (as little as 3 years and as much as 8 years), which are almost entirely advisor-determined. Explicitly seek out advisors that graduate students quickly. When meeting with potential advisors, ask questions like, “Where do your students go when they graduate?” and “How long do students in your lab take to graduate?” Also, talk with the professor’s current students to determine where they are in their progression and how long it took them to get there. Look on LinkedIn to find former students from the lab and to understand how long it took them to graduate. Former students can be found by looking at the lab’s history of publications. The first authors are typically the PhD students or post-docs.
In general, it is better to find older tenured advisors than younger advisors. Tenured advisors can make sure you graduate when they want you to graduate, and they have more control on your PhD committee. Younger, untenured advisors do not control your committee, so you are exposed to potentially more surprises.
Treat finding advisors like finding a job. There are great advisors and there are horrible advisors. Prepare a great CV. Send it to the professors that interest you along with a personalized note explaining that you’ve read some of their papers (which you need to actually read) and that you are interested in their research. Read their most current papers published by the more senior students. Trust me that other grad students are already trying to land the best advisors and get GRA-funded from Day 1.
Besides that, the benefit (and curse) of younger advisors (with less grant money and fewer resources) is that they will work very closely with you (or otherwise make sure they are riding you to get the most output they can), while older advisors with stable funding tend to be much more hands-off.
Follow Matt Might’s blog, read his posts about graduate school.
Follow Paul Graham’s essays, read the earlier ones about programming (like this one) and, if you’re interested, the later ones about startups.
Do not underestimate the PhD qualifying exam. You will probably fail the first time. Many do. I did. Work hard to prepare for that. It comes up in the first year of school. You can start preparing this summer (look online for the resources).
Explicitly seek out programs that pay better than the standard GRA salary. Look for IGERT programs which are default NSF grants. The Presidential fellowship is good too. When I was there, I averaged $35k/year in school funding, largely because I was very proactive in finding those higher paying grants and was lucky too.
TI:GER is awesome, pay attention to that towards the end of your 2nd year. The Sam Nunn Security program is awesome too.
Everything boils down to publishing. The more you write, the sooner you write, the better off you’ll be. Don’t delay getting published.
Classes are a necessary evil but have little to do with actually getting a PhD. Don’t worry about your grades any more. No one will care if you get C’s or if you get A’s. People will only care about how great you are at your specific area of research. Graduate classes tend to be easier than undergraduate classes.

Like anything else, there is a lot that can be optimized in the graduate school experience.

What advice to you give to incoming graduate students?

Related articles