I am PhD student in CS and I have a budget to employee a student assistant, but I'm not sure how I can manage them.

Unfortunately, my supervisor moved to another university. I'm still working with him, however, now I have no any another members in the research group here. The problem is that I have no idea what I can give a Student assistant as work. As I worked as a student assistant, usually I had my own research theme and took a part of research in the group. But there is no chance to enter new students to my research because I'm not a professor and I can't give a research theme as their graduation projects. Then I have no idea for what kinds of work that I can give them who can work for 40 hours/month. I think it's better to employee a student assistant if I have a budget though.

Well, I need to write some documents in the local language, so I think I can give it to a student assistant as work. But it's only one or two times for a year. Is there any tips to manage students who only can 40 hours/month?

Thank you very much in advance.

You're in a rare and fortunate but also challenging position, if you've not managed people before.

It is, in general, always better to spend budget in academia if you have it, otherwise it usually gets taken away.

One tip is to talk to the individual (or potential candidates), and try to identify their own long-term goals. If they want to do their own PhD, experience helping you analyse data or literature is something that they'll be more keen to do. If they want to go into industry, helping with programming aspects might be more useful to them. What's relevant will depend where you're up to in the PhD, but any boilerplate coding or statistical analysis etc. is fair game.

I would try to balance giving them tedious things they need to do, with interesting things they want to do. Bear in mind if it's just translation you need, you could pay a professional translator and probably get a better result for less cost. If they're intellectually-minded and/or ambitious, you may find they don't perform well at mundane, repetitive things; on the one hand, that doesn't mean you can't ever give them a mundane, repetitive task (that's why you pay them), but is also means you probably won't get the most out of them (and potentially a resignation) if that's all they do.

If you're stuck identifying interesting things they could do, ask them for their input (not doing this is a common management error). If you're interviewing, ask this to candidates. You may not be able to give them an accredited graduate project, but that doesn't mean you can't jointly devise something useful to you, and useful to them, and an employee will be more enthusiastic about undertaking a jointly devised task.

As a final note, the other thing to balance is giving them a task that's useful but not essential for your PhD. With any graduate RA role, you need a Plan B for anything mission-critical, since they don't exactly have a contract they're going to worry too much about quitting from, and limited experience/expertise.

I had a similar problem but with placement students, supervisor got an international placement student to come to the UK for the summer to help work on my project. Except, it was the first year of my PhD, I didn't have any work for him and she took a 10 week family emergency break. Lets say I failed miserably.

My advice, if possible make them do replications or side projects which you have a methodology for. I later got another placement student to replicate previous work with a different feedstock but with a very rigid methodology. If you can't give them a research theme, think of the tedious parts of your PhD that you hate doing and ask them to do 10X as much. It also helps if they are doing the same constantly as you have to teach them less and by the end of the time they are pretty good.