I actually just defended last December, so here we go: Yes, but I very much enjoy learning and tackling new problems. I might have been able to get the same amount of enjoyment in industry...
I actually just defended last December, so here we go:
Are you glad you spent all that time pursuing it?
Yes, but I very much enjoy learning and tackling new problems. I might have been able to get the same amount of enjoyment in industry somewhere, but I certainly wouldn't have had the flexibility to really pursue problems of my own choosing.
What does having one allow you to do that not having one would prevent you from doing?
The biggest thing for me is the amount of options I have available right now. I don't know that I couldn't have got there without the PhD, but it would have been much more difficult.
Do you still maintain connections with your advisor(s) and/or fellow students?
I still currently work at my university, so yes! I am also in touch with several previous graduates.
Are you proud of your research?
Very much so. I got very fed-up of closed source, in-house codes and the dubious results that the often produce. In addition to creating a code with good performance, I'm very proud that the codes I used, the data I created and even the scripts / LaTeX files I used for my papers are all open-source, and available for others to use or look at. I feel like that really should be a requirement in science, maybe we'll get there someday.
Do you still look at research in that field?
Yes, currently. I may be moving to a new job soon where I would be doing something related (GPU- and CPU-vectorized acceleration of scientific codes) but on a much broader scale, which I think is very cool.
What do you do now?
Currently I'm working on a project for my advisor involved with the automated generation of compact (i.e., one to say ten step) chemical kinetic models based on experimental data -- I finally get to really take a crack at applying some ML to my work, so that's interesting.
Do you get along with your advisor?
My advisor is very good to me, but also has high expectations for my work. I think that's a great balance.
How much time do you spend looking at publications in your field?
Some, not an exorbitant amount. I have read more publications from Comp. Sci., Math or even Electric engineering of late, simply because that's where the advanced integration / vectorization methods live.
Is most of your new knowledge from these publications, or do you perhaps rely on books you have not yet read?
I would say that a decent amount of my knowledge, especially initially, came from these publications. As I progressed however, I ended up more and more in a very specialized corner of existence where there really hasn't been much work. At this point, most of my knowledge came from things I learned myself. Once you pretty regularly start running into code issues that Google has no idea exist, you know you're off the beaten path :p.
How has your own funding (e.g. NSF fellowship) or lack thereof impacted what you do with your day?
I was lucky enough to (most) always have funding -- first through an NSF fellowship teaching STEM to high-schoolers, and later from a proposal I helped author. I greatly appreciate not having to be a TA, if only so I have more time to work on my own stuff.
What do you anticipate doing after you finish?
Right now it's looking like I'll either be doing numerical simulations at an aerospace company, or working on optimizing scientific codes for GPUs. Either would be pretty interesting to me.
What (open source?) tools do you find the most useful in your work?
Numpy, loo.py, pyopencl, matplotlib, POCL, and Cantera are all absolutely essential for my work. Of those, Numpy and matplotlib are probably the two with the broadest applicability.
How do you balance work/life?
Take breaks when you need to. Work twelve+ hours a day for a month only when absolutely necessary. Don't be afraid to just take a weeks vacation somewhere there in the summer, your work will be fine waiting for you.
If you are/were a TA, how did you learn how to be an effective one?
N/A
How do you make sure you are on track with your research goals?
This is tricky. I often way-overshot my timelines, specifically because the tools I worked with (OpenCL / CUDA, etc.) are finicky as hell. The important thing is to be smart in how you're making progress. If you get stuck, take a step back and rethink things rather than trying to keep charging forward. If you're lucky enough to have someone else working on a similar project who might be able to help (never happened for me) don't be afraid to ask them.
What are your biggest wins? Your biggest regrets?
My most recent code is ~33x faster than the open-source OpenFOAM CFD code for the solution of chemical kinetics. Some of that is probably because of OpenFOAM being slow, but still ... felt pretty good to see that number pop up.
As far as regrets... if you're working with code, LEARN UNIT TESTS. You'll be slower at first, but particularly for my work it was the only way to maintain sanity and make sure that things were actually working, rather than breaking in unpredictable and exciting ways....
If you have a PhD: Mostly, yes. I loved learning my subject, I loved teaching, and I loved research. Writing was frustrating at times but quite rewarding in the end. But the stress and money...
If you have a PhD:
Are you glad you spent all that time pursuing it?
Mostly, yes. I loved learning my subject, I loved teaching, and I loved research. Writing was frustrating at times but quite rewarding in the end. But the stress and money worries over so many years made it really tough. And the academic job market is awful so about halfway through I realized I likely wasn't going to have my "dream job" and worried I was putting myself through it for nothing.
What does having one allow you to do that not having one would prevent you from doing?
If you want to teach at a university level, of course you need a PhD. But in my field (anthropology) it also frankly gives you that prestige factor for when you are on the job market looking at applied/"real world" jobs. Having gone through a PhD also shows that you have the ability and skillset to 1) develop a research project 2) get funding for that project 3) independently conduct your original research 4) analyze data and construct a well-worded argument from it 5) present your findings. Note that this is specific to my field - in other areas students often have their projects more tightly controlled by their PI and their work is more collaborative. But either way, it is something on your resume that proves you're capable of that kind of work along with having developed skillsets applicable to a wide variety of settings.
Do you still maintain connections with your advisor(s) and/or fellow students?
I email my advisors periodically and meet up at conferences. I'm more in touch with my fellow students and have found them to be a very important part of my professional network. Honestly, professional opportunities have come more from them than advisors.
Are you proud of your research?
Yes. I did get to do original research on a cool topic that contributed to the field. It was tough to do and I think I came up with some creative contributions.
Do you still look at research in that field?
All the time, though I have a broader range now. Which, honestly, I like because I get to read what excites me and not just something for my dissertation.
What do you do now?
I work in science advocacy, communication, and engagement. I really like my job, it pays decent, and I feel like I'm doing something meaningful. I definitely couldn't have gotten this job or have much opportunity for upward mobility if I didn't have a PhD.
I just finished my MS in computer science and, in August, I began my PhD in data science and public health. Yes! I am co-advised by two people and I get along with both of them. Probably an hour...
I just finished my MS in computer science and, in August, I began my PhD in data science and public health.
Do you get along with your advisor?
Yes! I am co-advised by two people and I get along with both of them.
How much time do you spend looking at publications in your field?
Probably an hour or two every couple of days. Sometimes a lot more, sometimes not at all.
Is most of your new knowledge from these publications, or do you perhaps rely on books you have not yet read?
It depends -- certain types of knowledge doesn't come from either papers or books. A lot of it is experiential for me. I'm studying data science and public health.
How has your own funding (e.g. NSF fellowship) or lack thereof impacted what you do with your day?
It doesn't, really. I'm university-funded for six years.
What do you anticipate doing after you finish?
Ideally a professorship!
What (open source?) tools do you find the most useful in your work?
Basically 99% of the Python and R libraries I use. Numpy, scipy, etc. in Python and many, many things in R.
How do you balance work/life?
I refuse to work Saturdays and minimize work on Sundays. I also try to be done with work for the day before dinner, but that's not always possible.
If you are/were a TA, how did you learn how to be an effective one?
I've been a TA >20 times and a course instructor a couple of times. It's something that takes a lot of practice. I recommend attending workshops your institution might provide.
How do you make sure you are on track with your research goals?
Regular check-ins with advisors, PIs, etc. Work with your advisory structures to make sure you're meeting frequently enough and to make sure you're aware of what's expected of you.
What are your biggest wins? Your biggest regrets?
That's a tough one. My biggest win was finding the lab I work in. My biggest regret... I'm not sure.
Do you get along with your advisor? How much time do you spend looking at publications in your field? Is most of your new knowledge from these publications, or do you perhaps rely on books you...
Do you get along with your advisor?
Oof, ask me next week. I started a relationship with a woman in my lab. Her mom's cancer metastisized to her everything so she asked to take a leave of absence to take care of her...our boss decided that was not going to happen, despite it being allowed by our handbook. She talked to the department and quit. It's a whole thing, because she completed everything but her thesis. I'm currently sweating my rec letter and writing my thesis while supporting us.
How much time do you spend looking at publications in your field?
4ish hours a week.
Is most of your new knowledge from these publications, or do you perhaps rely on books you have not yet read?
Uh, yes. But there are books I..have read.
How has your own funding (e.g. NSF fellowship) or lack thereof impacted what you do with your day?
I spend 12 hours a week teaching and about 6 grading. It sucks, but my personal life has evaporated anyway.
What do you anticipate doing after you finish
Ideally, an academic position somewhere. I could go for a national lab, though.
What (open source?) tools do you find the most useful in your work?
Uh...Scidavis is a great alternative to Origin.
How do you balance work/life?
I don't.
If you are/were a TA, how did you learn how to be an effective one?
I had a great mentor for a semester. I teach an intro course, so the main goal is to make students explain their methods.
How do you make sure you are on track with your research goals?
Uh, my head? I just know where I want to go and keep going. I use a Hobonichi Techno because it has a page for each day, and it's a great planner/journal.
What are your biggest wins? Your biggest regrets?
any other things you want to talk about?
I got a nice fellowship for size months - I'd advise everyone to just send out write-ups, because it works.
My biggest regret is...nothing. I'm as happy as could be. It would be nice if I wasn't harboring a deep disgust with my boss, but whatever.
Wow, that is so incredibly fucked up. I'm really sorry that happened to her / you. GA's recently (in the last few years) won the right to unionize in my state. I was happy to give the Union some...
Oof, ask me next week. I started a relationship with a woman in my lab. Her mom's cancer metastisized to her everything so she asked to take a leave of absence to take care of her...our boss decided that was not going to happen, despite it being allowed by our handbook. She talked to the department and quit. It's a whole thing, because she completed everything but her thesis. I'm currently sweating my rec letter and writing my thesis while supporting us.
Wow, that is so incredibly fucked up. I'm really sorry that happened to her / you. GA's recently (in the last few years) won the right to unionize in my state. I was happy to give the Union some small chunk of my already small paycheck specifically to avoid shit like this (besides, the university would have found a way to raise my fees and make it go away anyways).
I have a PhD in STEM from the USA. I was often unhappy during my PhD. Assume it will take you 6-8 years to complete. Do not date within your department. Five years after graduation, you will...
I have a PhD in STEM from the USA.
I was often unhappy during my PhD.
Assume it will take you 6-8 years to complete.
Do not date within your department.
Five years after graduation, you will probably not use, and will have forgotten, your PhD work.
If you are unhappy after two years, I strongly encourage you to cut losses and leave with a free Masters degree.
I was on the path to get a master on literature but was unable to finish the degree because I was uneasy in a department that thinks the notion of truth is nothing more than western essentialism...
I was on the path to get a master on literature but was unable to finish the degree because I was uneasy in a department that thinks the notion of truth is nothing more than western essentialism and everything is an attribute of language. It didn't help that my advisor was a retarded cunt.
I actually just defended last December, so here we go:
Yes, but I very much enjoy learning and tackling new problems. I might have been able to get the same amount of enjoyment in industry somewhere, but I certainly wouldn't have had the flexibility to really pursue problems of my own choosing.
The biggest thing for me is the amount of options I have available right now. I don't know that I couldn't have got there without the PhD, but it would have been much more difficult.
I still currently work at my university, so yes! I am also in touch with several previous graduates.
Very much so. I got very fed-up of closed source, in-house codes and the dubious results that the often produce. In addition to creating a code with good performance, I'm very proud that the codes I used, the data I created and even the scripts / LaTeX files I used for my papers are all open-source, and available for others to use or look at. I feel like that really should be a requirement in science, maybe we'll get there someday.
Yes, currently. I may be moving to a new job soon where I would be doing something related (GPU- and CPU-vectorized acceleration of scientific codes) but on a much broader scale, which I think is very cool.
Currently I'm working on a project for my advisor involved with the automated generation of compact (i.e., one to say ten step) chemical kinetic models based on experimental data -- I finally get to really take a crack at applying some ML to my work, so that's interesting.
My advisor is very good to me, but also has high expectations for my work. I think that's a great balance.
Some, not an exorbitant amount. I have read more publications from Comp. Sci., Math or even Electric engineering of late, simply because that's where the advanced integration / vectorization methods live.
I would say that a decent amount of my knowledge, especially initially, came from these publications. As I progressed however, I ended up more and more in a very specialized corner of existence where there really hasn't been much work. At this point, most of my knowledge came from things I learned myself. Once you pretty regularly start running into code issues that Google has no idea exist, you know you're off the beaten path :p.
I was lucky enough to (most) always have funding -- first through an NSF fellowship teaching STEM to high-schoolers, and later from a proposal I helped author. I greatly appreciate not having to be a TA, if only so I have more time to work on my own stuff.
Right now it's looking like I'll either be doing numerical simulations at an aerospace company, or working on optimizing scientific codes for GPUs. Either would be pretty interesting to me.
Numpy, loo.py, pyopencl, matplotlib, POCL, and Cantera are all absolutely essential for my work. Of those, Numpy and matplotlib are probably the two with the broadest applicability.
Take breaks when you need to. Work twelve+ hours a day for a month only when absolutely necessary. Don't be afraid to just take a weeks vacation somewhere there in the summer, your work will be fine waiting for you.
N/A
This is tricky. I often way-overshot my timelines, specifically because the tools I worked with (OpenCL / CUDA, etc.) are finicky as hell. The important thing is to be smart in how you're making progress. If you get stuck, take a step back and rethink things rather than trying to keep charging forward. If you're lucky enough to have someone else working on a similar project who might be able to help (never happened for me) don't be afraid to ask them.
My most recent code is ~33x faster than the open-source OpenFOAM CFD code for the solution of chemical kinetics. Some of that is probably because of OpenFOAM being slow, but still ... felt pretty good to see that number pop up.
As far as regrets... if you're working with code, LEARN UNIT TESTS. You'll be slower at first, but particularly for my work it was the only way to maintain sanity and make sure that things were actually working, rather than breaking in unpredictable and exciting ways....
If you have a PhD:
Mostly, yes. I loved learning my subject, I loved teaching, and I loved research. Writing was frustrating at times but quite rewarding in the end. But the stress and money worries over so many years made it really tough. And the academic job market is awful so about halfway through I realized I likely wasn't going to have my "dream job" and worried I was putting myself through it for nothing.
If you want to teach at a university level, of course you need a PhD. But in my field (anthropology) it also frankly gives you that prestige factor for when you are on the job market looking at applied/"real world" jobs. Having gone through a PhD also shows that you have the ability and skillset to 1) develop a research project 2) get funding for that project 3) independently conduct your original research 4) analyze data and construct a well-worded argument from it 5) present your findings. Note that this is specific to my field - in other areas students often have their projects more tightly controlled by their PI and their work is more collaborative. But either way, it is something on your resume that proves you're capable of that kind of work along with having developed skillsets applicable to a wide variety of settings.
I email my advisors periodically and meet up at conferences. I'm more in touch with my fellow students and have found them to be a very important part of my professional network. Honestly, professional opportunities have come more from them than advisors.
Yes. I did get to do original research on a cool topic that contributed to the field. It was tough to do and I think I came up with some creative contributions.
All the time, though I have a broader range now. Which, honestly, I like because I get to read what excites me and not just something for my dissertation.
I work in science advocacy, communication, and engagement. I really like my job, it pays decent, and I feel like I'm doing something meaningful. I definitely couldn't have gotten this job or have much opportunity for upward mobility if I didn't have a PhD.
I just finished my MS in computer science and, in August, I began my PhD in data science and public health.
Yes! I am co-advised by two people and I get along with both of them.
Probably an hour or two every couple of days. Sometimes a lot more, sometimes not at all.
It depends -- certain types of knowledge doesn't come from either papers or books. A lot of it is experiential for me. I'm studying data science and public health.
It doesn't, really. I'm university-funded for six years.
Ideally a professorship!
Basically 99% of the Python and R libraries I use. Numpy, scipy, etc. in Python and many, many things in R.
I refuse to work Saturdays and minimize work on Sundays. I also try to be done with work for the day before dinner, but that's not always possible.
I've been a TA >20 times and a course instructor a couple of times. It's something that takes a lot of practice. I recommend attending workshops your institution might provide.
Regular check-ins with advisors, PIs, etc. Work with your advisory structures to make sure you're meeting frequently enough and to make sure you're aware of what's expected of you.
That's a tough one. My biggest win was finding the lab I work in. My biggest regret... I'm not sure.
Ideally, an academic position somewhere. I could go for a national lab, though.
any other things you want to talk about?
Wow, that is so incredibly fucked up. I'm really sorry that happened to her / you. GA's recently (in the last few years) won the right to unionize in my state. I was happy to give the Union some small chunk of my already small paycheck specifically to avoid shit like this (besides, the university would have found a way to raise my fees and make it go away anyways).
I have a PhD in STEM from the USA.
I was often unhappy during my PhD.
Assume it will take you 6-8 years to complete.
Do not date within your department.
Five years after graduation, you will probably not use, and will have forgotten, your PhD work.
If you are unhappy after two years, I strongly encourage you to cut losses and leave with a free Masters degree.
I was on the path to get a master on literature but was unable to finish the degree because I was uneasy in a department that thinks the notion of truth is nothing more than western essentialism and everything is an attribute of language. It didn't help that my advisor was a retarded cunt.
Now I'm in IT.
That is reasonable, and also very different from the irrationality I faced in the literature department.