Dear Prospective PhD Student, Pay careful attention to the optics of this advice. It is addressing how you can best serve your PI. Your goals and aspirations are not addressed. It is advice on how to land a job, but does not address why you should want the job, or if it is a good fit for you.
Perhaps you are a good fit for the academic life. But seek out more advice on what the job of Prospective PhD Student is. Seek out more advice on the job market of jobs that require a PhD. If you are thinking of working in the university setting, there may be a lot fewer tenure track jobs, with a lot more applicants than you might imagine, available that you.
IMO you want to distinguish between advice that is tactical and short term or strategic and long term.
The best tactical advice tends to from people who were recently in a very similar position as you, had had approximately the same goal you have, and either achieved the goal or failed. You want a sample of both successes and failures. The less recent, the less useful the advice.
The best strategic advice tends to be from people who have accomplished things you want to accomplish, regardless of time distance. Their tactical advice is less useful (because it is often dated), but the strategic advice tends to be better. You again need a large sample size to filter out noise.
You trust no one, and, at the same time, everyone. Determining what you are getting yourself into, based on data, not your imagination of the jobs based upon your undergrad experience, should be your first research project. If you aren't up for doing that research, evaluating the data you gather, and seeing the bias in the data, a research position probably not a good fit.
It should be a mutually beneficial relationship. That said, I run into far too many PhD students who want to join my lab who cannot answer the question "Why do you want a PhD?"
I work in AI, so there is enormous opportunity inside and outside of academia currently (was not true when I finished my PhD, when it was the least popular area of CS), but I'm very upfront with students about how challenging it is to get a tenure-track job. That said, for those who want that, I help them make plans and set goals to achieve that objective, which means having a much stronger CV by the time they graduate than for those aiming for industry.
Many may disagree in practice, but I see my primary job as creating strong scientists and helping them achieve their career objectives. But, I won't take PhD students if their goals would not greatly benefit from having a PhD.
I do think too few academics are upfront with students about how challenging it is to get a tenure track position and what's needed to pull that off.
I usually have only 1-3 slots per year. It is a lot of very hard work. If someone can't explain why they want a PhD, they are likely to not complete the program and drop out, in my experience. It takes enormous resources in terms of my time and grant money to train students so I want them to thrive and graduate.
It’s foolish to spend 5-6 years getting paid subsistence wages because you wanted a PhD for fun. Also, phds are not really fun. Grad school is much more miserable than pleasurable.
Strangely my path both supports and counters your statement. Likely in no way typical. But given the prevalence of negative comments regarding grad school in response to this article I want to say:
I thought getting my PhD was fun and a great time in my life.
(looking back with 30 years of hindsight)
Start to end my PhD was a roughly ten year period.
Round 1: (in support of "no fun") Had a great undergrad experience (thank you Harvey Mudd). Entered grad school with goal of teaching at an undergrad only liberal art college. Frequently read Chronicle of Higher Education so I knew there was an over supply of PhD to tenure track positions at the time, but I was special, so no problem. Passed qualifying exams. Received great advise and directed away from my area of interest because it was already full of tenured faculty and was told, correctly, that departments were so full up in that area and it would be near impossible to get a tenure track position in the sub area of CS. I started thesis, published, even had expanded version of a paper invited for inclusion in book. I was playing by the rules, but seeing how the sausage was made, I was ground down and quit
Intermission: (An aside, I was always amused as an undergraduate how the janitors would enter classroom and empty trash can during ongoing classes. So my first post drop out job was a version of that). I got a job as sysadmin in the department I had just dropped out of. Got to go to former professors offices to fix their networking problems, change printer paper for them. Then former classmate was starting a company and I got hired by member of my qualifying exam study group to be first hired employee and company he was co-founding. After quitting that job, I was visiting a friend back in Claremont who was busy inventing MIME. Mentioning that I was unemployed, he said HMC Academic computing was looking for new User Support and Education Coordinator, the former one having quit just before school started. Walked me over to Director of Academic Computing's office. Half hour later I had a job. (Not quite janitor, but closest to cross it off bucket list). Being back in the environment rekindled the dream that sent me to grad school in the first place. So I reapplied to grad school
Round 2: (not following the rules, but having lot of FUN). The solid advice, one shouldn't pay tuition to go to grad school. Me, I paid tuition out of my own pocket, received no institution support as RA or TA. Basically doing it all wrong. My original advisor, in immeasurable kindness, agreed to again be my thesis supervisor. While providing no financial support, I was still a time sink for her. There were moments of pure joy when I got to be a non conformist grad student (aka pain in the butt). Like when the department secretary sent out a letter to all grad students asking help moving furniture from point A to point B. Instead of being a helpful department servant, I got to fire off an angry letter (reply, faculty include, all of course) saying that I was paying my own tuition and wasn't there to provide free labor. I had fun because I was in control of my time. I had fun cause there where great recreation classes on campus. I had fun cause the local vegetarian cafe was the default punk music venue of the city. I have fun cause I was doing things on my terms, studying in a field that I love.
Postscript: Despite odds, got a tenure track position and was assistant professor for 3 year. (Reality was not the same as dream.) After that got another job, again through a (different) member of my qualifying exam study group. Approx 8 year after receiving PhD, I was more or less retired, having easily achieve the financial goal I established as a grad student learning how to enjoying life on a frugal budget.
I feel like this article is missing a gigantic point: your PI is everything as a PhD student, if you have a PI who’s a shitty person (sometimes even unknowingly), you are going to be in hell for four to seven years. Don’t pick your PI just based on «they seem to be at the forefront of research on this field », also check how they treat their students. This is not like a normal job where you can easily quit, you’re likely going to be stuck with your PI, so they better be good.
This 120%. Your day-to-day, your mental health, and your research success are largely a function of your lab's culture, and it's the PI who sets that culture. And the unfortunate reality is that PIs really don't get any formal training on culture-setting and people management, so there are some really successful folks out there who suck at running their groups.
The way you tilt the odds in your favor for getting a good PI-student match is to be embedded in the field. If you're an undergrad, getting involved in research is a great low-risk way to do this. People in research have reputations in their field, and people in the know will direct promising students away from labs with bad cultures.
1. Don’t opt for a phd just because it sounds cool (remember the first tweet shared in that article).
2. Academia is a pretty hard space if you can’t get your own funding or if you don’t like hustling for it. Getting tenure track jobs without having funding attached is impossible.
3. Very few jobs in industry need a PhD. Most are ok with a masters. Those that require a PhD can be selective based on where you graduate from and what your publication record is.
4. Know your PI well and reach out to his/her previous students who have already graduated. Some may reply - especially those with a bad experience.
5. Back to point #1, don’t take up the PhD just because you get the assistance and stipend. Make sure you really like your subject - it’s the only thing you’re doing for the next 4-12 years depending on your discipline while subsisting on poverty level stipends
Any job that is highly research oriented is going to appreciate a Ph.D.
Whether it becomes a practical requirement, or formal requirement, would depend. If it's a high paid position, in a small research group, of an established field, then Ph.D.s are likely to be very important.
If its a position in a large research group in a new and fast changing field, then its likely to be more open to anyone with unusual or interesting promise, but still highly value Ph.D's.
As for fields: research in engineering, finance, economics, device physics, chemistry, biology, pharmaceuticals, etc. Anywhere the research involves highly technical knowledge, and a pre-existing track record of original thinking, personal initiative, that have produced objective results, are very important.
"Research" is a thought and project management meta/leadership skill, on top of being highly educated and talented in an area. Ph.D.'s provide the extra time and social context to fill out those skills and demonstrate a track record.
In any of those fields, and computers science, there is a spectrum.
Deep research areas in computer science would be foundations of databases, cryptography, formal verification, statistical heuristics, core algorithms behind deep learning, ...
Since so many areas in computer science are relatively new, there is generally more flexibility in who might be considered great at research. The field includes many unusually creative individual practitioners who have produced great work without a Ph.D.
But Ph.D.s are still very much in demand and common for research positions in these highly technical, high value areas.
Robotics is one such field. I work with tons of PhDs, and self driving car companies hire them like crazy. In certain organizations, you need a PhD to advance as an Applied Scientist. A masters won't cut it.
But that’s my entire point. Taking your example, the bulk of robotics companies aren’t all applied scientists (for that matter in spacex, nasa, etc.). You will have a very small cohort that does the r&d and the rest of the folks executing on that.
The r&d space is really interesting but very selective for obvious reasons. Just a PhD won’t get you in. You will need significant other contributions - eg the right publications in high impact journals, possibly internship with the companies themselves, the right advisor/PI that collaborates with the company, etc.
The team I'm on is half PhDs. We aren't doing fundamental research, we are building a real product. We prefer to hire PhDs as applied scientists (which we distinguish from research scientist) because they are better able to apply and build new-ish ideas from academia. But most of what we do is not novel in the publishable sense. The PhDs I work with are certainly all smart, but aren't the top echelon you are describing.
Not the GP and I agree with your broader point that a master's is sufficient approximately 99% of the time, but a few places come to mind: SpaceX, NASA, CERN (is that technically non-academic? I don't truly know), probably certain departments of US Dept. of Energy, things around nuclear power.
They probably don't have high concentrations absolutely (e.g. they're not going to be 90%+ PhDs) but it's probably still several orders of magnitude above your average industry role.
IC design for behemoths like Intel/AMD/Nvidia. AFAIK none of those is gonna let you anywhere near their multibillion-dollar design without being at the top of the field or having lots of experience (>5yrs) already
Is 5 years really the bar for "lots of experience?" I feel like this is a byproduct of current (last 10 years or so) frontend/bootcamp dev mentality that you're a senior engineer after a 6-week course and 2-3 years of experience at a consulting body shop.
Even if you get a PhD your working career is going to be around 35 years. Add 5-10 if you're done after college, and even more if you're not doing college.
I think we need to stop pretending people are senior when they're 10-15% of the way through their career.
Yes - this is unfortunately everything. There are a number of categories to optimize (research area, lab size, funding, university, other lab members, etc) but the biggest factor is your relationship with your PI. I cannot imagine working as a PhD student in a lab where I don't deeply respect and get along with the PI. I was very lucky but selected for this when choosing PhD labs. A surefire way to have a terrible few years is to ignore others advice about PIs to avoid.
I chose mine because he was a reasonably guy to work with and talk to. What I did not know at the time was that he was very junior in the department and needed to please other professors. So even though I liked working with him, I was quite helpless at certain difficult parts of the process.
Far too few incoming students value this, even if explicitly told. They will go to the lab of the most famous professor at the highest rank university, even if that PI is not a great mentor. Not all PIs are intentionally aiming to make their student's lives hell, but many inadvertently do by providing insufficient mentorship, which results in students suffering huge amounts of anguish because they are struggling to excel and know it.
If a PI isn't meeting with each PhD student for at least 30-60 minutes per week, especially in the first few years, they are probably not doing a great job.
I always encourage students to email the current and former students of the professors they want to work with once admitted to see if they can have a video chat to discuss what it is like working in the lab. Even then, it can be hard to get frank feedback.
This is my advise to people who ask me for advise when applying to PhD / graduate school for research. The advisor is probably the only other main factor besides finances and any other personal constraints. If you have a great advisor all around (not just the best researcher in their field, as in they also know how to mentor, teacher, network, and help you reach your goals), they will know how to navigate your different research interests and goals of your PhD and get you to where you want to go.
I would just flat out advise against it. Don’t set your life up for a dice roll on which PI you get. It can be good. It can be bad. But it’s not a wise gamble for any aspect of life quality that I think is particularly wise to orient towards.
* talk to your PI beforehand obviously, so you can see if they are a jerk
* keep an eye on the failure mode. Check if you are making real progress after a year or two. If not, mastering out in a STEM field a totally legit path that will still leave you with a perfectly decent career potential. Some places will even let you do a thesis with your masters if you for some reason love writing giant LaTeX documents.
progress can be extremely nonlinear - my year or two progress check would have yielded little, but ultimately I was in great shape when graduating my PhD program. Your advice isn't bad, but it's helpful to understand that the foundation can take a while to lay and then progress can accelerate rapidly towards the end.
This is definitely true. I've had some students who ramped up quickly in the first year or two, but some of my best students didn't ramp up until year 3 and had no output before then. It was very stressful for them, though, since students are prone to comparing themselves to the very best.
Yep - exactly. I'm not sure if there's any way to know if progress will accelerate or not. Personal research breakthroughs are likely a combination of preparation, luck, and environment. That can require patience, although I can say that at no time did it feel like my situation was hopeless or way off track. The conditions were appropriate for a breakthrough. If I hadn't seen the light at the end of the tunnel, I think that might have been a signal to switch tracks.
as someone who mastered out SEVERAL years into it this is the only advice anyone will ever need regarding a PhD. of course if you come from money and can roll a dice on a career go for it
As someone that runs a PhD program in economics, I'd like to see "in a grant field" added to the title. It's completely different in fields with limited grants (those that hire TAs rather than GAs). One of the things I wasn't expecting when I started this position is the number of students that use advice from the sciences/engineering as if it applies to all fields. (And a lot of my email time is spent fixing those incorrect beliefs.)
> Most PhD programs in science and engineering will come with free tuition, a stipend, and health insurance.
Don't go if it doesn't come with funding. There should be a tuition waiver (not always a full waiver, I didn't get one), but...
> fees typically aren’t. This can still be a significant amount of money each semester
This really is ridiculous. They can be in the thousands of dollars. Make sure you subtract fees from the stipend. You might need to pay taxes on the full stipend, but not be able to deduct the cost of fees (I don't know current law).
> Again, program stipends will vary widely. Some may offer a stipend, but not guarantee it past 1-2 yrs.
This is a concern if you're funded by a grant. I know lots of programs that don't guarantee five years of stipend, because they aren't allowed, but in practice they'll do everything they can to keep funding you if you're making progress. That's because completed PhD's is a big deal when evaluating the research status of universities.
> health insurance. Most programs will offer graduate student health insurance. But, as with stipends, the monthly premiums and quality of insurance can vary widely.
If you can get insurance through your parents, you almost always want to go that route. I was surprised when I saw how much this varies from school to school. My employer covers 75% of the cost (one of the highest I've seen) but it still costs the student $700/year. As with fees, subtract the cost of health insurance from the stipend, but only after you determine the quality is sufficient for your needs.
Yeah I’m currently in my 5th year in a top Econ program and a lot of this is just foreign to my experience. The torturous part was how my department handled Covid but after that it’s been pretty smooth sailing.
My advice is find a PhD you want in Norway or the Netherlands. They have programs with competitive research groups, salaries that will get you a quality of life you will appreciate, they will end in 3-4 years, and they have money to pay for you to go to conferences, etc without you having to fight for it from your adviser. Everywhere else sucks.
This comment is based on my experience working as a scientist in academia in USA, Japan, Germany, France, Norway, UK, and now the Netherlands.
Add Luxembourg and Switzerland to that list - at least EPFL and ETH. It's so unbelievably better to be at an institute where travelling within Europe is simply not a budgetary problem at all.
Your experience at ETH might vary a lot though. Their PhD programs are extremely heterogeneous between departments and vary from modern, well structured grad schools to the old-fashioned central European style where the student are basically at the mercy of their advisor for five years.
Currently the salary is not competitive however Poland is the shining star of Europe given their growth and everything else these days. I might go there if science funding becomes more inline with other European countries.
When I was an undergrad, I once asked one of the department professors "I want to do a PhD, how do I do that?" and the first thing he said to me was "Don't."
It turned out to be pretty good advice. I didn't in the end, and having seen close friends do it, I'm glad I was dissuaded.
Since it seems just about everybody who's been in or around a PhD program agrees it's a miserable institution with low chances of satisfaction, I will continue to follow the advice that it is not for me.
However, are there alternative paths to a research career? I really enjoyed being in an academic environment in undergrad and dream of working on AI/ML research in some capacity, but I'm a total outsider.
> it seems just about everybody who's been in or around a PhD program agrees it's a miserable institution
I disagree, quite strongly in fact. But my experience is not with US PhD programs, which is what discussions here often center upon.
As for working on research outside of pursuing a PhD: one option is to become a scientific programmer. Typically, you'd work on helping to execute the research, without the pressures of having to publish.
Of course, your "clients" do have that pressure, but your job typically is to Make It Happen, explain why it can't, and find crafty workarounds to Make It Sorta Happen Anyway.
Getting a PhD is hard, and it certainly does take a mental toll. Students are trying to push up against the boundaries of human knowledge to expand humanity's knowledge. That isn't easy, and because we train them to be independent researchers, much of that time is spent working on solo projects with their mentor supervising.
That said, once one has a PhD, it really opens up many exciting careers that are otherwise not attainable. If one wants to be a scientist and do research, you almost certainly need a PhD. It also gives you so much more autonomy in terms of the projects you work on rather than just being a coder working on someone else's projects. I only have one life, and we spend much of our lives working. I'd rather have the ability to control what I work on and wake up excited to get to work every day.
for yourself, write down in great detail what your dream is. What do you imagine yourself doing day-to-day? What experiences did you really enjoy?
Identify (via posting here, linkedin, discord) some people that have the sort of job you imagine and have a dialog with them. (if they are actively posting, they likely have the time to converse with you)
If your exerpience is anything like mine, you may find the activities you enjoy are a small part of the researchers jobs, which in many cases is a grant writing machine.
Having seen what my partner has been through in her PhD in generics in the UK. I can certainly advice against doing it.
PhDs are not worth the stress your will be through. You are barely paid to get by. PhD students are exploited left and right.
The whole concept is not suitable for what's life in 2023. If you have a toxic colleague at a workspace you can do something about it, change team, change your manager, change workspace.
If you have a toxic colleague or supervisor it's done, your will be miserable for 4 years. Then if you need extension suddenly fees will pop up.
I'm seeing way too many comments about not being able to change the relationship.
I was in a top school. I saw plenty of students change advisors and do fine. You have to be a bit careful - your former advisor shouldn't be in your thesis committee, and some advisors are unwilling to take you if they think it will cause bad blood between him and the original advisor. But it definitely was done. Over and over again.
Of course, it helps if the department is big (mine had about 100 professors).
They are worth it if you’re trying to emigrate out of a bad situation. My partner is in a lab with an awful PI, but she is Lebanese, and wanted to be able to bring herself and her family out of a very unstable situation.
Does it ever make sense to do graduate school just to get by? If you’re desperate for work, does the stipend ever cover the cost of living, at least enough to spend all your time in the library?
Just come to Norway. As a PhD student you're actually a full time "federal" employee with pension benefits, you're entitled to sick leave, parental leave, etc. You have 5 weeks of paid holidays per year. Your starting salary is $49k per year, which is decent - about the same level as a nurse or a teacher.
To be clear: if you're a woman who move here with your partner for a PhD, and you get pregnant during your PhD, you will be entitled to 9 months of fully paid maternity leave, plus your planned dissertation date will get postponed by 9 months, no questions asked. You can even do this twice if you're "efficient". Kindergarten is quite cheap at around $320 per month. As a man, I had one kid before starting my PhD and the second about halfway through, then I got 10 days leave just after the birth, and 3 months leave from our daughter was 9 to 12 months.
There are several bona fide unions you can join as a PhD student, you have a relatively well functioning HR system that will help quite a bit against toxic PIs. You will have annual "evaluation talks" with a professor other than your supervisor, where you can voice any concerns you might have. The PhD programme duration is 3 years, maybe some people do an additional few months, but I've never heard of someone doing more than 4.
It's still hard work though, there is supposed to be blood, sweat and tears. But it should be because sciencce is hard, not because you are living in poverty or because you have to postpone starting a family until you are in your fourties.
For some reason we get a lot of candidates applying from Europe and Asia, but almost nobody from North America, and I don't see why.
As always, "it depends" but it's very possible if you go with the understanding that you're going to live like a poor college student for the next half decade and don't want or need your expenses to go up like your peers that go into industry.
You also have to take into account that you're going to be working more hours than a traditional job and that you're going to be stuck there for half a decade, but also you get something out of it at the end which you won't get just working some other low pay job.
I'm seeing a lot of answers in the negative, so I'll give my experience:
Find a relatively low cost of living area. The good news is that there are several top universities in such places (e.g. Midwest).
Find out the pay you'll get.
Find out the fees you'll pay.
Now you know how much you'll make. Check rents in the area, and then decide.
For me, (2003-2010), as a single person, the pay was more than sufficient to get by. I could afford a single bedroom apartment, got a good used car, and drove across the country (staying at motels, not camping) twice. I bought a DSLR, and had a very nice PC that lasted me all those years. Would eat out from time to time.
It wasn't much, but it was all I needed, and a bit more.
The main difference between then and now is that cars have gotten a lot more expensive. Rents may have gone up, but perhaps not as much (i.e. pay may have gone up enough to compensate).
Back in 2010, studying for a PhD in the UK meant I got an EPSRC stipend of £13500 per year. I was fortunate to find rooms to rent in houses with other grad students for around £350-400 per month. I didn't go out much, cooked my own food, commuted by bike, and made use of free or low cost uni services like gym and therapy. On some weekends I drove ~100 miles to visit my girlfriend, but I was a pretty lousy partner at the time and didn't do that as much as I wish I had. Also petrol was a fair bit cheaper back then. The only time I remember coming close to worrying was when my stipend cheque got put in the wrong place and I didn't get it for a few weeks.
>Does it ever make sense to do graduate school just to get by?
most people who do them are coerced -- i made about as much as my pizza delivery driver. i attempted it because it seemed a way to become financially independent. instead, when i tried to leave with a masters i was forced to eventually return to my homeland and be subjected to the same things that drove me from it.
rarely sufficiently covers the cost of living, though yes in some cases (e.g. if you get the top end of the stipend range from a rich university that also happens to be in a low-cost-of-living area).
terrible idea for someone who just wants to work
(if you can get into a good grad school you can find some job that pays better and has better working conditions); conceivably not a terrible idea for someone with specific career goals who is trying to ride out some specific industry downturn.
The one advice I often do not see is: While there may be enough jobs for PhDs out there, there are very few that you will want to do.
People get a PhD because they are passionate about the field and want to do research. Not just any research but research on a few topics. Most jobs for PhDs will not let you do the research that you would like to do. It will be totally on you to target those few jobs that will. You have to work towards it throughout the program. Actively network at conferences, and get key people to know you. Position yourself for those jobs. Merely doing good research and publishing papers may not be enough.
Anyone who sticks to it can get a PhD. But no one feels sorry for a PhD grad who can't get a job. It's expected that if you're smart enough to get one, you're smart enough to figure out the job situation.
The PI on the only research project I ever worked on as a student was relatively chill and collaborative, so I assumed that's how they all were. 20 years later, I once again work with big research projects, and was surprised to learn how tyrannical and petulant PIs are (or have a reputation for being). Much more so than CEOs in any private sector company I've worked at, the PI seems to be both at such a high level that they are disengaged from the daily work of their "employees", and simultaneously inclined to micromanage that work when their attention turns to it at last—the worst of both worlds!
Best advice for quality of life for a PhD student living off the PhD stipend - live in a low cost of living area. In math (what I got a PhD in), you get paid the same amount (about $20,000 a year when I started in 2013) if you live the SF Bay Area (sat going to UC Berkeley) as you do if you live in Tucson, Arizona (say going to University of Arizona). There are good universities in low cost of living areas that prospective PhD students should heavily consider - at least I wish I did.
If you live in the US, have the ability to work in the US, and have a stem degree, the opportunity cost of a PhD is just way too high, many hundreds of thousands of dollars. Technically, one could do the PhD in the UK where PhD is 3 years, to minimize that opportunity cost, but not by much.
You should not get a PhD for the money. That's always been the case. It's a research degree. If it will open doors for the type of work you want to do, that's the only time it's worth it.
Traditionally, the reason to get a PhD was that it was the only way to get an academic job. In 2023, I don't recommend academic employment. Sadly, it's just another type of employment today. I started my first academic job in 2002 and things were different back then.
What do you feel has changed in the last 20 years? (This has been discussed a lot but my impression is that people compare today to the 1960s more than to the 00s).
It's not a 0 dollar difference, that would be doing something not for money, but a huge negative difference is doing something for negative money. Bad idea IMHO. Life isn't a video game; you only get one play thru.
Somebody has to do it. We wouldn't have the civilization we have if we had no scientists. Similarly, we need teachers, caretakers, stay at home parents, and somebody has to repair the phone lines and pave the roads. If every person on the planet simply pursued the maximum pay career available to them, we wouldn't have much of a society at all.
You should certainly be aware of the financial sacrifice you're making to do socially impactful work and consider this in the decision calculus, but I don't think you can say zero people should ever make that decision.
Yes, unless that PhD puts you on an earnings track where you can quickly make up for a few lost years of wages and saving. Otherwise, the compounding effect of having missed investing a few hundred thousand dollars _at a young age_ is substantial.
First off: SW is an outlier. The opportunity cost is a lot lower in most other STEM fields. And for a few, the net wealth is actually more with a PhD (see what jobs you can get with just a biology/chemistry undergrad, for example).
It is problematic to view it purely from the perspective of money. Consider the huge non-financial opportunity cost to not doing a PhD. If you get a PhD, you still have a chance of founding a business and becoming a multimillionaire. If you don't do a PhD, you'll almost certainly not get the benefits of the PhD ever.
I look at my friends who went straight to industry after their undergrad, and compare with myself and those who did a PhD, and there's a notable difference. If your advisor is not a slave driver[1], you'll have more time to introspect and be exposed to many perspectives/disciplines, at high concentration, than you have time for with a full time job. You can do it after work, but only if you opt not to have a life. It will takes perhaps 10 years of consuming all your non work time to perhaps match up 5 years of grad school. Pretty much no one goes this route.
Put another way: If your son discovered he had more earning potential if he never completed high school, would you say "Go for it!"?
[1] And contrary to what you read on HN, many are not.
A couple of years ago I would also have said that a PhD is not worth the opportunity cost. I look back to the the time when my wife and I were both fresh out of college and in our PhD programs. We lived _very_ frugally. After housing, food (all cooked ourselves, no take-out or eat-out), transportation (had to live outside of the city), we had maybe $100-$200 left over each month (in 2000-2004 dollars). We had no money to set aside for retirement or investments. We did take out interest free student loans and put them in a CD and paid them back immediately after we graduated, so that helped provide a little investment income. We worked hard to scrape by on a poverty-level stipend without going into debt.
Shortly after we got real jobs, I looked at our friends who went to industry right after college and they were significantly better off, and continued to be better for a quite a while (that whole compounding rewards things).
However, now that I'm senior in my position (real senior, not SV senior label), I have much more stability in my job than my non-PhD friends have. The PhD is pretty valuable for the work I do, and that keeps me as the lead on most of our contracts.
I can't say if I would be better of right now if I went to industry right away and started making money and savings, or if the stability has put me in a better place.
I don't have the massive income as an SV tech job, but I have also never been cut at a layoff, so I don't have to worry about the variability in income.
As an aside, I will say that a PhD is non-trivial and you should probably only consider it if you _love_ the field. If you're just using it to buy yourself some time or to live off the stipend because you can't find work, you probably won't have a fun time.
This is mediocre advice that largely won't work for the members of this community. For neuroscience, where there is relatively little competition, it's ok. For AI/ML, where the competition is crazy, it's bad advice.
> Lastly - the dreaded Undergraduate Publication. Are you a co-author on a publication? Fantastic. It means you made a considerable contribution to a project which can only be done through a significant amount of dedication. Are you first author on a publication? Wow! Truly impressive and a rare feat for an undergraduate researcher. Do you not have any publications? Don’t sweat it. I’m looking for people with a history of a strong work ethic, an innate curiosity, and an ability to think and dream big.
Downplaying this, and cutely calling it the "dreaded" publication is the worst advice on this page by far. This is what gets you into good departments now. A publication as an undergrad can change your life. If you want to do a PhD, this is your goal as a student: publish a paper.
The majority of new PhD students who get into good departments now have at least one publication. And what you did for that publication is what matters. Your supervisor(s) should include in their letter exactly what your contribution was.
In our department the guiding philosophy for admitting PhD students mostly is: Are they already behaving like PhD students on the level of those we have in our department? A publication or more is a really good signal. The cleanest and strongest signal you can send.
> There are two primary ways most undergraduates get research experience. One, a summer Research Experience for Undergraduates (REU). Two, working with a PI at your home institution for a few semesters. Neither is required for admission, but both are valuable for different reasons.
I can hardly imagine a student getting in to our department with zero research experience at all aside from exceptional situations. These are definitely required.
> First zoom meeting ... Generally, I’ll try to keep this first meeting focused on you, your interests, your long-term goals, and whether my lab fits those interests and goals. Somewhat counterintuitively, I’ll actually avoid going into depth about science. Of course, we will talk about my research program and open projects in my lab, but I really try and keep this first conversation focused on getting to know you (and letting you get to know me). I want to know how you got to be where you are today, how you made the decision to pursue a PhD, and where you see yourself going in the future.
You must read the last few papers of the person who is calling you. You must be ready to talk about all of your work end to end, not just about what you did, but its limitations, what else you wanted to do, other approaches, etc.
> For neuroscience, where there is relatively little competition
Insane take. Neuroscience is highly competitive. As someone in both fields, I would say it is only a little bit less competitive than AI/ML.
I largely agree with the rest of your advice, but I would argue that having multiple extremely positive reference letters from different research experiences are worth more than some middle author publication.
this is sound practical advice but reflects a bad system that can't be sustained.
you kind of have to be Junior Patrick Bateman now. students scheming and fighting for UG research opportunities. this level of overachiever striving used to be reserved for people aiming for investment banking or McKinsey.
it can be done and the students who do it can be very productive and actually do good research. but it's also now filtering for a very different personality type; less "curious" and more "do anything to win". not that the latter necessarily makes a bad scientist as long as they stay honest. but a field where almost everyone is like that will feel very different.
ultimately, I guess, it is just supply and demand. it's in departments interest to only admit students who they are 100% sure will produce a lot of papers for 5-6 years no matter what the department does. and they can do this because a lot of people want to come.
As is frequently the case here, I feel most comments about whether or not to do a PhD are way too absorbed in cynicism and negativity.
Sure, a bad PI can make you miserable, and if you only catch on when you've already invested serious time into research, you might be stuck with that PI until you're done. But, at least from what I've seen at my university, you have a good amount of flexibility during the first years of the program. Plus, after the first semester, our department expects PhD students to do teaching assistant work if not funded by other means, which gives them the chance to work with professors without getting tied down to a PI, making it easier to make a more informed decision.
It's worth emphasizing that if you really love the topic, have confidence in your skills and can tolerate your PI, it can be a very fulfilling albeit stressful experience. My PI is typically well spoken, very demanding and often an asshole but also gives enough freedom and autonomy to grow. The pay is also definitely nowhere near as good as it would be in industry.
But as someone who would consider himself to have been pretty immature and mentally 'weak' (struggling to cope with anything stressful that wasn't an exam or project) when starting the program, I feel I have grown significantly as a person over the past few years in a way that would have likely involved a lot more exploitation and pain in industry (especially given my immigration status, which also tends to encourage exploitation). When I started, I struggled even with basic 'adult' things, yet 4 years in, approaching the end of my program, I am able to operate in the group mostly autonomously, with just occasional progress updates and handling basic decision making - reducing my PI's workload etc.
Additionally, as someone who had always been interested in physics, but pursued an education in computer engineering since that was my stronger skill, a PhD ended up giving me the chance to participate in cutting edge physics research despite my computer specialization.
I may have lost out on some wealth, but I feel I've more than made up for it through the personal growth, the satisfaction of unintentionally fulfilling a childhood dream, the satisfaction of making a measurable and meaningful contribution to science and the significantly expanded immigration opportunities compared to the conventional painful H1B->Green card route for Indians.
I can't speak too much about admissions, as my experience with that was fairly unusual - I transferred in from a Masters at the recommendation of the department chair with most other requirements waived because they remembered my classwork from undergrad and felt a PhD would fit me better. I was also set up with an introduction at a lab to fund me (granted that this did mean I didn't get any flexibility in choosing my PI). But I disagree with the general premise that PhDs are not worth it.
> Non-academic jobs and experience - should I include it?
> I am a huge supporter of two things not related to research and education: having a job while an undergraduate or showing a substantial dedication to an organization.
Sadly, this is a very ageist perspective, that I think is very common in academia. This advice is specifically geared towards undergrads who are applying to PhD programs without stating so. It's fine to tailor your advice to certain audiences (in fact the post says it's the author's perspective), but at least state who the intended audience is. I think the reason this ageist persepctive is common in academia is two-fold:
1) Many academics came straight from undergrad (or maybe had a one/two year gap), so they reflect their personal background as indicative of the path others will take to academia.
2) Older students typically have significantly higher self-esteem and understand their worth. They've potentially achieved success outside of academia where their talents were appreciated. It's much harder to steer such students to eschew their self-worth.
I speak from personal experience here. I started my PhD program at 35 after a successful industry career where I was unhappy with the type of work I was doing. I realized after a terrible interview experience at OpenAI that a PhD was the only way to pursue my interests in the burgeoning AI/ML/NLP field. Luckily, I made it into a pretty good PhD program that I'm happy with. I now conduct research melding LLMs with video games an interest I've held since the 90s playing CRPGs (back then I didn't have a specific technology in mind, just the idea of an "AI narrator").
If I had to give advice to PhD students, it's to know your worth. The fact that you made it into a PhD program is not a fluke. Imposter syndrome is rampant and many advisors take advantage of this, either explicitly or implicitly without realizing it.
I came from the video game industry, so I knew crunch times, but my first year as a PhD student was exceedingly brutal. To get my first paper out, I spent three weeks working 100 hours a week. I walked out of the lab the day before the paper deadline and nearly quite my PhD program. I just thought nothing is worth this stress.
If my wife didn't take care of everything else during that time, I couldn't have done it. During that exact same time, I was taking classes. I had a three week project that coincided with my paper deadline. I didn't start the class project until the day after it was due (you lost 10% of your grade per day late). Having industry experience helped tremendously here: I was already quite familiar with distributed systems having been the lead engineer on a (never released) MMORTS.
Don't do this to yourself. It's not worth it. At least in CS, deadlines for conferences are pretty arbitrary. Good research doesn't fit into fixed timelines.
As an aside, I don't blame my advisor for my stressful first year. He became a professor the same year I started my PhD. He didn't know any better and just assumed I was managing my time well while getting my work done. We had a good talk after the deadline and he's been absolutely great since. That's why I say, sometimes it might happen out of pure ignorance of the system. No one teaches you how to be a good advisor while you're getting a PhD. You're assumed to learn that once you become a professor.
Though there are those who DO take advantange knowingly. There's a professor in our department who is an absolute nightmare and for each candidate weekend when new prospective students are deciding on programs, all the current CS PhD students actively warn them away from working with this individual. They quite literally work people to the brink of death at times, where people have been hospitalized for exhaustion, and the professor even contacted them to do work while they were in the hospital! Though, considering the one suicide in their lab, it's debatable if "brink of death" goes far enough.
I stopped reading the original post and the comments here as my brain stuck at PI, what the heck is that? Not all the readers are PhDs and PI is 3.14 or Raspberry to me these days.
Principal investigator. i.e., the person in contractually in charge of a research effort, often used as a shorthand for a supervising professor who heads up a lab and advises PhD students.
The only advice that a prospective PhD student needs to have is telling them to stay the hell away from a PhD and never even touch it in their dreams with a ten feet pole.
Almost completely unrelated to the article, but I figure a few of you prospective PhD candidates could use a mental break anyway, so here goes.
Had a nice Baader-Meinhof moment from the article. I have been obsessed with music from a young age, not just the output but also the creative and technical processes that come together to make it, yet despite reading and watching a lot about my favorite bands, the creative process, the music industry over the years I've never internalized his name (although I've doubtless encountered references to him many times).
I stumbled onto a documentary about Rubin and his studio Shangri-La last night and found it immediately captivating, musing to myself, "Wow, what a cool and weird dude. How have I never picked up on his influence before?" After the doc finished I enjoyed a nice session of creative thinking inspired by some things he and David Lynch were talking about. Cool, I've got a new person on my radar to seek out their words and wisdom when the mood or need arises. Then in the opening paragraphs of the first link I click this morning, on HN in a piece written for PhD candidates of all places, here is his name. Kind of wild.
The song "Hurt" by Johnnie Cash [0] is a long-time favorite. It's one the greatest rock and roll covers ever, in no small part for how transformative and emotionally powerful the reinterpretation is, second only to "Take It to the Limit" by Etta James [1] imo. I just learned from that interview it was Rubin who brought Reznor's song to Johnny, saying "this is you, this is the man in black." It's funny, I have sometimes wondered about Johnny listening to Nine Inch Nails, finding it weird. But this is somehow even crazier, that Rubin could hear that first song and somehow connect it to Cash, knowing he might be able to turn it into something so powerful and personal.
I feel like anyone who has a fair degree of creativity in their pursuits, whether professional or for personal fulfillment, can benefit from absorbing some of this dude's process, thoughts, and vibes. Even if you find him and what he says completely ridiculous. Anyway, the entire 60 minutes interview that the clip linked in the article is from is worth a watch [2]. It's pretty short. I particularly like an exchange from the end of that interview,
Rubin: "The audience comes last."
Cooper: "How can that be?"
Rubin: "Well, the audience doesn't know what they want. The audience only knows what has come before."
In the brief time I've been exposed to the guy, one of my favorite concepts Rubin expressed is this idea that his best creative ideas come entirely from outside himself, so most of his life is about putting himself in a place where he can be receptive to that outside force or influence, whatever it is. That rather than being a generator of great ideas, he is merely an antenna that can be carefully tuned to receive the signal from somewhere else.
We all have heard the wisdom that you really can't force creativity or flashes of insight, but this takes it a step further in a way that resonates with me. I don't think what he is saying is necessarily true in any objective or empirical sense of course. The power of this concept is that it deeply acknowledges that almost everything inside us that we think of as ourselves, especially our ego, our sensory interpretations, and our conscious thought processes pouring through our knowledge base, are nowhere near enough on their own to produce the kind of brilliance which human beings are sometimes capable. And becoming overly focused on those more superficial (or at least obvious) parts of our consciousness can sometimes drown out or distort the signal that comes from deeper awarenesses which sometimes have unique and powerful things to say about reality.
Thanks for reading, and good luck to you in your creative and technical pursuits!
Perhaps you are a good fit for the academic life. But seek out more advice on what the job of Prospective PhD Student is. Seek out more advice on the job market of jobs that require a PhD. If you are thinking of working in the university setting, there may be a lot fewer tenure track jobs, with a lot more applicants than you might imagine, available that you.
EDIT: https://www.nature.com/articles/d41586-022-03394-0