[bmqib]

I'll be starting an undergraduate degree in Computer Science, and I was wondering if anyone would suggest what I could do from right now to make a good PhD application, and how to make sure I have funding etc. sorted by the end of my undergraduate degree?

My further studies will be in Computer Science (machine learning/bayesian inference) too...

[The_Lonely_Goatherd]

One step at a time! Get the grades for your undergrad, start your undergrad, see how things go and whether you still like studying by the time you hit the end of your third year

A stellar transcript will always help though

[kikkoman]

(Original post by bmqib)
I'll be starting an undergraduate degree in Computer Science, and I was wondering if anyone would suggest what I could do from right now to make a good PhD application, and how to make sure I have funding etc. sorted by the end of my undergraduate degree?

My further studies will be in Computer Science (machine learning/bayesian inference) too...

Good grades (and win some prizes), try and get summer research studentships. Maybe even get your name on a publication as 3rd or even 2nd author if you are lucky. The stellar references will come along naturally.

However, you haven't even started university yet. As the poster above me said, one step at a time. Concentrate initially on getting a First in your first year and take it from there

[poohat]

(Original post by The_Lonely_Goatherd)
One step at a time! Get the grades for your undergrad, start your undergrad, see how things go and whether you still like studying by the time you hit the end of your third year

A stellar transcript will always help though

I dont agree with this, its always good to be planning for the future especially if you want to maximise your chances of getting into a top place

The 3 things that matter for PhD applciations are undergraduate grades (+ institution quality), research experience, and letters of recommendation. If you can do some work alongside a lecturer/postdoc (eg summer research experience) then that would look very good. Similarly, if your university has any courses which are based around projects, take them all. That way you will meet as many academics as possible, who can give you research experience and write you references. Realistically if you work with an academic on a project you are likely to be a glorified code monkey until you get more experience, which is fine. Make sure your programming skills are rock solid, and it will be easier to find people to work with you.

Computer science is unique among academic disciplines in that most research is published in conferences rather than journals. Conference papers tend to be significantly shorter than journal papers and are in some sense easier to publish (except the top tier conferences). Therefore its not unrealistic for you to get a publication during your undergraduate years if you can work with someone good; while this isnt necessary for a PhD application, it does look very nice on your CV.

[poohat]

Actually from skimming your post history I agree with The_Lonely_Goatherd after all, you need to just calm down and relax. Yes, its nice to be prepared and so on, but stressing out about whether you are good enough for a million completely separate things before you even start university doesnt make much sense. Just chill out, do your first year while learning as much as possible, do the society thing so you have extra-curriculars for the private sector, and take any project/research based courses that are available. But dont stress yourself out worrying about every little thing.

[Exulted]

Hello bmqib,

When I was your age, I haven't even thought of doing a PhD .

This year I wanted to do a PhD, but, unfortunately was unsuccessful at 3 unis.
But, I learnt valuable lessons from my mistakes with which I will share with you:

1) You should start your PhD application well in advance. I got my MSc certificate in early
January and made my 2 applications in early February. It was too late.

2) You should contact your supervisors before you do your formal application.
I knew about that, because my classmate who is doing her PhD at Leeds uni told me
about it. But, I decided first to contact admissions department at the uni and their PG
administrators advised me to submit the formal application without contacting potential
supervisors. I just lost my time waiting in vain. I missed the deadlines for scholarships at
most of the unis after I got my rejections from that 2 unis.
3) Maybe, you have to ask your referees to send you their references. By doing so, you
won't annoy them to write the references on your behalf numerous times. And you will be
more relaxed, too. Find out whether unis accept such references (whether it doesn't
violate their data protection act).
I wanted to do everything right, and asked my referees to send directly their references
every time. Of course, it was too time-consuming for them and nervous for me.
Moreover, right now, I don't have any of their references in hand.

4) Then, right now, all you need to do is study hard and get excellent grades.
5) I think doing a Masters degree before your PhD would be advantageous. It will definitely tell you whether it's worth doing a PhD.

But, because, you are just starting your undergraduate degree, I advise you not to over analyse your future life. Everything's changing so quickly.
And I'm sure your needs and wants will change after 3 years.
Now, during your undergraduate degree, you have to check whether Computer Science is your ultimate passion for which you would like to devote considerable time in academia.

Enjoy your first degree and have a good time!

[Ghost6]

I have already seen very valuable information in this thread. As the other posters pointed out, grades and recommendations will be crucial. Nobody mentioned graduate coursework so far, but having some of it on your transcript will definitely look good and it will be one more thing for your recommenders to comment on, provided you do well. You could go as far as getting a master's if you wish to crack the top 10. You will then have at least one year of graduate classes under your belt and some significant research experience through a master's thesis of some sort. The master's is particularly important if you attended an average undergraduate institution and your CV is lacking in the "prestige" department. Generally speaking, top master's are much easier to get into than PhDs, because they come unfunded. If you will already have strong to very strong profile at the end of your undergraduate years and are looking for a competitive edge for a top 5 to top 10 program, the master's could be definitely be worth the cost and the effort. Plus, it will make the transition to the PhD softer. And if your PhD applications fail or you decide against going for a PhD, a CS master's is a valuable qualification for many jobs as well. The downside is obviously the cost and the fact it will delay your PhD applications by two years, because the objective is having the master's in hands for your PhD applications or at least one year completed if it lasts two years.

[bmqib]

(Original post by Ghost6)
I have already seen very valuable information in this thread. As the other posters pointed out, grades and recommendations will be crucial. Nobody mentioned graduate coursework so far, but having some of it on your transcript will definitely look good and it will be one more thing for your recommenders to comment on, provided you do well. You could go as far as getting a master's if you wish to crack the top 10. You will then have at least one year of graduate classes under your belt and some significant research experience through a master's thesis of some sort. The master's is particularly important if you attended an average undergraduate institution and your CV is lacking in the "prestige" department. Generally speaking, top master's are much easier to get into than PhDs, because they come unfunded. If you will already have strong to very strong profile at the end of your undergraduate years and are looking for a competitive edge for a top 5 to top 10 program, the master's could be definitely be worth the cost and the effort. Plus, it will make the transition to the PhD softer. And if your PhD applications fail or you decide against going for a PhD, a CS master's is a valuable qualification for many jobs as well. The downside is obviously the cost and the fact it will delay your PhD applications by two years, because the objective is having the master's in hands for your PhD applications or at least one year completed if it lasts two years.

By top 5-10 you mean worldwide right?

Also is there any way you know of to get research work experience in my first year? I'll see if I can work with my uni's lecturers, but do you think it's likely I could do this with people at other universities?

[The Lyceum]

At first year you're not really relevant in any sense, let alone assisting research, even in something quanty like Comp Sci. Try for internships.

I have a friend in Comp Sci and his CV is pretty ****ing amazing, essentially what makes it great is: Grades, he's easily a first class candidate often venturing into the 80s. Secondly, he did one or two internships at labs during his summers in his first and now second year. His plan is to graduate with a high first, with some sort of research project in his third year and then go on from there.

Do something like that in your first two years.

[kikkoman]

(Original post by The Lyceum)
At first year you're not really relevant in any sense, let alone assisting research, even in something quanty like Comp Sci. Try for internships.

I have a friend in Comp Sci and his CV is pretty ****ing amazing, essentially what makes it great is: Grades, he's easily a first class candidate often venturing into the 80s. Secondly, he did one or two internships at labs during his summers in his first and now second year. His plan is to graduate with a high first, with some sort of research project in his third year and then go on from there.

Do something like that in your first two years.

A friend of mine did exactly that. Mid-high firsts every year, 1 month research internship from 1st to 2nd year, 2 month Wellcome Trust funded project from 2nd to 3rd and a 2 month project at one of the research institutes in Berlin. Needless to say, he easily got on to a top programme.

[DynamicSyngery]

Factors of importance in order:

1. Degree class. Get a first.

2. Recommendation from your final year project supervisor or whoever (should be someone who actually knows you, not a generic cover letter from the director of studies).

3. Research experience in the summer.

Each is dramatically less important than the one preceding it, or at least the later ones are worthless without the earlier ones. In the US I think research experience is more important, but here many people do not have the opportunity to do it and it's not so common in general, so there's less discrimination on that basis.

[bmqib]

Thanks, really appreciate the posts... I was wondering if someone could advise on funding, anything that could help with that?

[flying plum]

(Original post by bmqib)
Thanks, really appreciate the posts... I was wondering if someone could advise on funding, anything that could help with that?

I think a lot of the advice above is just as relevant to funding. Getting a place on a PhD programme isn't really as hard as you'd think, with some exceptions; but getting funding is another challenge altogether. Not getting a first and distinction doesn't preclude you from funding (I have neither), but it certainly makes it a hell of a lot easier. If you have weaker grades, relevant work experience and a clear reason for choosing and love for your subject/field certainly helps, and a track record in research is definitely helpful, weak or strong grades.

You need to keep in mind that funding for Master's degrees is even harder to come by than PhD funding, so if you are thinking about doing postgraduate studies, you might want to start thinking about how to fund them come the summer of your second year, as many funding deadlines are in the January preceding the start of the academic year.

[bmqib]

One final question, I know this is repeated over and over again, but in the likelihood that I miss my grades and go to a lower ranked university (QMUL), what kind of disadvantage would I be in? Would it make it that difficult to get on a good MPhil+PhD from a 'bad' university?

[flying plum]

(Original post by bmqib)
One final question, I know this is repeated over and over again, but in the likelihood that I miss my grades and go to a lower ranked university (QMUL), what kind of disadvantage would I be in? Would it make it that difficult to get on a good MPhil+PhD from a 'bad' university?

If you get excellent marks on undergrad, probably not. It certainly wouldn't preclude you from getting a place for a PhD, but it might be harder at 'top' Universities (particularly if you want to go internationally for a doctorate). But I'm no comp. sci. person, so I couldn't really tell you if it's different in that field.
And who knows what will have happened in the three/four years it takes you to complete your undergrad.

[Cirsium]

(Original post by bmqib)
One final question, I know this is repeated over and over again, but in the likelihood that I miss my grades and go to a lower ranked university (QMUL), what kind of disadvantage would I be in? Would it make it that difficult to get on a good MPhil+PhD from a 'bad' university?

In a way it could actually work in your favour. Internships and summer jobs are relatively hard to come by, because the money for them has to come from somewhere and there isn't usually spare cash sloshing around in grants. Assuming that you continue to work at the top of your game at QMUL, rather than getting complacent around other people who aren't as academically gifted as you, then you might be in a more competitive position to get those summer jobs, which are a major deal for applications.

You mentioned summer jobs etc at other universities. A lot of these will be internally advertised only, but there are various research council sponsored things e.g. in biology the Gatsby Foundation funds some summer projects and a summer school. Bearing in mind that computer science is presumably less expensive to actually do than wet lab stuff, you may be able to get work experience by asking if you can shadow someone without being paid, if that's feasible with where you live at the moment.

[bmqib]

(Original post by flying plum)
If you get excellent marks on undergrad, probably not. It certainly wouldn't preclude you from getting a place for a PhD, but it might be harder at 'top' Universities (particularly if you want to go internationally for a doctorate). But I'm no comp. sci. person, so I couldn't really tell you if it's different in that field.
And who knows what will have happened in the three/four years it takes you to complete your undergrad.

Not interested in any places other than in the UK because they take too long, but do you think it's going to be a problem for Cambridge or maybe Oxford? (I know there are other things to consider etc. when choosing which place to do a PhD in, but just curious...)

[Cora Lindsay]

(Original post by bmqib)
Not interested in any places other than in the UK because they take too long, but do you think it's going to be a problem for Cambridge or maybe Oxford? (I know there are other things to consider etc. when choosing which place to do a PhD in, but just curious...)

The best place to do a PhD is not necessarily what is simplistically perceived to be a 'top' university. Quality of research in particular area is very dependent on the individual academics, and the best prospective supervisors are not always where you might expect them to be. To give just one example, some of the UK's best condensed matter physicists have recently moved from Cambridge to Queen Mary, London so, if that's what floats your boat, Queen Mary is the place to go.

[Ghost6]

(Original post by Cora Lindsay)
The best place to do a PhD is not necessarily what is simplistically perceived to be a 'top' university. Quality of research in particular area is very dependent on the individual academics, and the best prospective supervisors are not always where you might expect them to be. To give just one example, some of the UK's best condensed matter physicists have recently moved from Cambridge to Queen Mary, London so, if that's what floats your boat, Queen Mary is the place to go.

Still, I would advise the OP to go to Cambridge if in that situation, especially if he ever wants to get a job outside of the UK or some non academic position. Cambridge is ranked third in the world (after Harvard and Stanford) for "brand name recognition".

[flying plum]

(Original post by Ghost6)
Still, I would advise the OP to go to Cambridge if in that situation, especially if he ever wants to get a job outside of the UK or some non academic position. Cambridge is ranked third in the world (after Harvard and Stanford) for "brand name recognition".

I'm not sure. I agree with Cora in terms of PhDs - who you're working with is more important. If you're in some specialist branch of physics, to use her example, en other academics will know that Uni of x is known for that, even if joe bloggs - or even prof joe bloggs in another academic field - don't really know the institution. For undergrad...I'm not sure. I know plenty of people doing masters at good institutions with undergrads from unis I'd never heard of (Uni Zagreb, universities in Kazakhstan etc), so although instinct suggests that a 'brand' such as Cambridge certainly carries kudos and might make it easier, how much of a disadvantage not going there makes is something, not being at a high enough level of academia, I'm not sure I'm placed to gauge.