HACKER Q&A
📣 move_the_sun

Bridging the gap between formal physics education and research?


I finished my undergraduate education in physics and computer science not too long ago, but have been working in FAANG ever since. My passion has always been physics but unfortunately to pay off loans/medical bills/etc I was more or less forced into software engineering.

I took the standard undergrad physics 'core curriculum' in college (e.g. Griffiths quantum, e&m, K&K mechanics, etc). I've been keeping up with my studies since then by slowly working my way through various 'graduate texts', e.g. Sakurai's text on quantum mechanics and Kardar's on statistical physics of particles.

I'd like to do some side work in an experimental quantum computing lab and have been in contact with various universities; however, what I'm finding is that there's a large gap between the material in the papers being read and written by the groups I've spoken with, and my formal education. The papers are constantly referencing ideas, equations, and experiments that went unreferenced in all the texts I've read thus far. Trying to Google and understand these ideas and experiments one-by-one from Wikipedia and articles has been inefficient and ineffective so far for me.

Moreover, since I have no university affiliation, it seems like these groups are not particularly motivated to help me bridge that gap; I have not received much in the way of advice in terms of getting up to speed.

Has anyone else experienced and/or solved this problem? Ideally I'd like to be able to (a) understand at a high-level the current state of the field, since there seem to be a myriad of different research directions all aimed more or less at the same objective, and (b) understand open problems for at least one of these directions (e.g. open problems in quantum dots or ion traps).


  👤 Donthatme Accepted Answer ✓
Here are my thoughts, as someone who completed a physics PhD not too long ago and switched academic fields.

There are two parts to your question: 1) How do I study up on advanced topics efficiently and 2) How do I do research with my current circumstance.

For 1), I would first continue working through grad-level textbooks. If you need some structure look at the textbooks used for required course at your favorite/dream grad program/school. These will hopefully give you a solid foundation of the physics/math for current research.

The next part about specific papers is tricky, because a lot of physics (sub-)subdomains are highly specialized and have their own terms and notations. In general though, my process is to look at the references and citations, and sometimes the references and citations of those. I am basically either looking for a paper or two that frames it in a way that I have seen before, or points to an old textbook that I am not familiar with. The process takes a bit of time at first, but is smoother after doing it a few times.

For 2), it depends on how deep into research you want to go. The current path would lead well to being well-versed in the current landscape as an outside observer. I think in all likelihood to do anything more than that you will need to have some affiliation with a research institution. If you want to be able to pursue your own research interests within those institutions, you will very likely need a PhD. Unfortunately it is hard to give more specific advice without knowing more details/desires. If you are in the bay area, you could also look at LBNL or LLNL and see if there is anything interest there for you.


👤 abdullahkhalids
This is not going to be easy.

1. Undergrad courses are definitely helpful but they are baby stuff compared to research level study.

2. Taking relevant grad courses on OCW might be helpful but less helpful than you think. For instance, my phd is in theoretical quantum optics. And I took a quantum optics course in grad school. I don't think I ever used anything I learned in that course in my research. Physics grad courses are often precisely designed to work in that prof's group, and not useful otherwise.

3. But you will have to take those courses anyway to up your base knowledge level and mathematical skills.

4. Read papers and work backwards and learn the stuff mentioned in the background sections. Do this for 6 months for few dozen papers. At which point you might to start to understand roughly what one paper has to say.

5. Open problems are often mentioned in papers and talks by field leaders - because they want others to build on their work.

6. But honestly this is not going to be easy. What you are trying to do is working at the home gym alone - except no one would ever see your muscles. PhD students working in labs feel lost. I can't imagine doing this on your own.


👤 tgflynn
Of course the traditional answer to this question would be to do a Ph.D. in physics.

If that isn't an option there are usually more specialized books in specific research areas beyond the level of undergrad/beginning grad-level texts. For example if you search for "quantum dots" on Amazon you will get a number of hits.

That said a significant part of learning a field of research is becoming familiar with the primary literature of that field (ie. published papers). In fact if you were in a Ph.D program you would probably also have to spend quite a lot of time and effort reading papers to get up to speed.

If you encounter an equation you don't understand there is undoubtedly some paper where it was first introduced. If it's widely used there may also be review papers which attempt to give a broader understanding of that equation in context. Usually finding these is a matter of doing a literature search (Google Scholar is quite useful for this) and tracing references in papers once you've found them.

Have you tried asking your contacts to recommend a list of books and key papers in their field ?


👤 Jugurtha
>Moreover, since I have no university affiliation, it seems like these groups are not particularly motivated to help me bridge that gap; I have not received much in the way of advice in terms of getting up to speed.

Could bringing your software engineering skills to these groups be your way in? Specifically, developing software for them, automating, version control, experiment tracking, environment setting, reducing the "I lost a month worth of work because I spilled coffee on my computer/double clicked on an article with .exe extension" type of issues many people in academia encounter.

I believe that there's a quantum leap in research that could be achieved just by using version control, let alone having someone like you on the team.

In other words, reduce their toil and overhead and accelerate their research. Once you're really useful and bringing value, the conversation changes. You can therefore leverage your existing skillset.


👤 jimmyvalmer
What academia does to pay the bills (publishing papers to get grants) has very little to do with undergraduate curricula. Once you're on "the other side" as a PhD student, you'll see undergrads as cattle whose primary function is to pay tuition.