About the Speaker and the Talk

George Skretas is a postdoctoral researcher at the Algorithm Engineering Group at Hasso Plattner Institute. His research deals with both static and dynamic networks, studying their behavior in centralized or distributed settings. After having studied computer science at the University of Patras, he obtained his PhD degree in 2022 from the University of Liverpool. While the first two talks of the lecture series focused on explaining how the scientific world and research works, this talk answers the questions of how and why to become involved in research, which usually starts with doing a PhD.

How to Become a PhD

The formal process for obtaining a PhD title is clearly defined by laws and regulations of the degree granting institution. However, the application process and daily work as a PhD student varies drastically depending on the institution, the group, and the research topic of a candidate. Thus, the talk presented an overview of possible arrangements.

Finding a PhD Position

The first step towards acquiring a position as a PhD student is finding an opening to apply for. While it is possible to apply for advertised positions, the scientific community is highly word-of-mouth driven and not every position will be published or might only be published to hire a specific candidate. Hence, it is recommended to ask departments or groups whether they have an opening and apply for a position that has not been published yet. Groups and researchers at HPI can offer guidance on whom to contact and reach out to students they have worked with previously. It might even be possible that a research group is willing to open a new position for applicants whom they want to join the group, although this requires a good reputation based on the applicant's previous work.

Various communities also have tools that can be used to look for positions. For theoretical computer science, there is the DMANET mailing list, but websites like cstheory-jobs.org or findaphd.com can help as well. Of course, these tools should not be trusted blindly. Word of mouth is still an important indicator to catch any red flags, so it is a good idea to ask around and talk to current and former students. Two characteristics to look out for are the group's publication record and the group size, as larger groups are more likely to suffer from poor supervision.

Formal Requirements

After finding a position to apply for, one usually needs to hand in their application. Again, different groups have different application processes, but diplomas of the candidate's high school graduation and their bachelor studies will always be requested. Since some countries, such as the UK, accept PhD students who have not completed a master's degree, submitting a master's diploma will neither be a requirement in those cases nor if the applicant has not yet completed their master studies. Depending on the group, the applicant might also be asked to submit a curriculum vitae, detailed grades and a list of publications, as well as reference letters or contacts for possible authors of reference letters. While good grades, especially in subjects related to the research topic of the position, are important, they can be trumped by publications, research projects and teaching experience, since these tasks are highly relevant for PhD positions. A good reference letter can more than make up for bad grades or missing publications, whereas a bad one can burn even a candidate with good records from other documents. The letter provides employers with someone trustworthy, such as a supervisor who has already worked with the applicant and can confidently judge the applicant's skills and potential.

Interview Process

Usually, obtaining a position requires passing an interview. The process for this is, again, highly individual. It might include a customized test at home or the task of preparing a presentation of a paper to see how the candidate works and how well they can understand new research, as this is the main task of new PhD students. Technical questions, live coding tasks, and questions on the motivation for a PhD are also to be expected, depending on the research field. Questions aiming towards the PhD plan of the candidate, topics they would want to work on and possible researchers of the group to collaborate with can be used to gauge the interest of the applicant and align expectations. In some cases, the candidate will also be invited to visit the group, this often signals high interest of the group in working with the candidate. Another possibility is that reference letters are asked for after the interview, although at this point they are more of a sanity check than anything else.

Statistics on the Application Process of Current PhD Students

An application might not always turn out successful, but applicants that show the necessary dedication will eventually find a position that fits them. An anonymous survey among PhD students at the Algorithm Engineering Chair shows that a majority of them were successful with their first or second application. Only a small number of people had to apply more than five times before getting their PhD position. Although there is some selection bias from looking at those who were successful, this still shows that one should not give up after the first application is rejected. There are numerous examples of people who had success at a later point in time.

Furthermore, the survey shows that many rejections are silent, i.e. applicants do not receive any answer to an application. It also emphasizes the fact that getting a position by word of mouth is quite common, as there are multiple PhD students at HPI who got their position without ever applying. This applies mostly to students who were in contact with their research group during their master studies, which is why attaching to a group early on during one's studies is a good practice.

Everyday Life of a PhD Student

The everyday life of a PhD Student will also vary heavily depending on their field of work and the group they work with. However, there are some common truths that hold for most PhD students in computer science.

PhD Contracts and Teaching

A PhD student is usually either employed by a contract or receives a stipend. A typical salary ranges from €1500 to €3000 net income, where the lower end is usually set by the stipend, which tends to be lower than a regular PhD contract salary. The lifecycle of either a contract or a stipend varies in different countries. A typical contract lasts for at least 3 years, but few finish their PhD within that time frame. In the end, an extension is always possible, but this heavily depends on the ability to secure funding for the position and whether the chair is willing to extend the contract or stipend. While it is possible to complete a PhD without being employed at the granting institution, this is rather rare, at least in the field of computer science.

A PhD contract usually specifies a certain amount of teaching duties. It is important to note that regardless of the teaching hours stipulated in the contract, a supervisor might ask for more. This is then usually not up for negotiation, leading to a non-negligible potential for exploitation.

Research

Especially in the starting phase of a PhD, students spend a lot of time understanding their research field and catching up on current research, meaning they spend a lot of time reading papers and less time working on their own findings, this ratio shifts as the PhD progresses.

While PhD students are often employed by a research institution, they still mostly work for their own progress and advancement, all while their research output is easily comparable to that of other researchers by publication count. This ease of comparison, along with the fact that a perceived lack of progress can last for months at a time, contributes to the wide spread of imposter syndrome among PhD students. All together, this can create quite a stressful environment, especially when compared to other office jobs.

Further, PhD students are often expected to self-organize and manage their time on their own. While a good supervisor will support them in setting a direction for their research, students still need to decide for themselves what they want to do and read, as well as whom to talk to and work for. This gives them a lot of freedom to plan their schedule but requires a certain level of discipline. The talk summarized the level of freedom in time management with the quote: "I am in academia, so every day is the weekend, and we work on weekends."

Aside from the large freedom that comes with a PhD position, another benefit of working in research is getting to present one's work at international conferences, traveling to interesting locations and meeting other researchers. All in all, the talk concluded that mental stamina and self motivation are some of the most important qualities required to successfully complete a PhD.

Receiving the PhD and Beyond

While life as a PhD student can be described as "conducting research" to a large extent, graduation and thus escape requires writing a thesis. In order to hand it in, some requirements must be met. Depending on country and discipline, publications may not be formally required, however, they are necessary in practice. For instance, graders will expect elements of the thesis to have been published and thus passed through the peer-review process. As a rule of thumb, a reasonable number of publications might be one less than the number of years spent as a PhD student, although four papers are sufficient in any case.

Assuming these requirements are met, and the thesis defense is successful, what comes next? Most graduates work in industry, even though circumstances need to work out for a PhD to help with this career path. For the most part, when the position does not match the previous specialization, having the title is neither a positive nor a negative. For this reason, pursuing a PhD for the sole purpose of getting a higher salary in industry is likely ill-advised. On the other hand, a PhD is absolutely necessary for a career in academia. Almost all long-term positions, including teaching, require the degree. After graduation, people often gravitate to new topics, something that is expected to some extent in order to demonstrate independence. The decision of whether to pursue a career in academia depends not only on the PhD student's research focus, but also their personal needs regarding predictability and job security, as a career in academia often implies fixed-term contracts and frequent relocations.

Why to Get a PhD

With a better understanding of what getting a PhD entails, should master students apply for a PhD position? That naturally depends on their motivations. As cautioned before, money should not be the primary motivator, since industry salaries exceed those in academia in almost all cases. Similarly, pursuing a degree for the prestige afforded by the title sounds pretentious, and the high regard for academic titles in Germany is far from universal in the world. On the other hand, having an affinity for teaching or interest in academia are good reasons to consider a PhD. The master thesis resembles a microcosm of the life of a PhD student, so enjoying that work is a great indicator as well.

The experience of obtaining a PhD can be described in a few words: creative, unstable, stressing, and independent. There is a lot to learn, and the experience will surely improve one's self-management. In the end, the purpose of a PhD is to improve and progress, so any time spent is not lost, even without completing the degree.

As illustrated, the road to PhD can take on many different forms, depending on the research area, the research group and of course the PhD student themselves. Especially taking the diversity of possible tasks and positions into account, it might be worth it to give a PhD studentship a go if it seems like a viable career option.