General information#
Applies to any kind of students projects.
Finding an supervisor#
As an assistant professor, I have a teacher and researcher role. As we mostly teach BEng students in the Engineering Technology department, I prioritize BEng students in my supervision. As a researcher, I want to carry my research field forward, so BSc and MSc students are also very welcome, as long as your project/company is appealing for my research.
Ideally your internship area & final project topic should be in my area of industrial expertise – FPGA design, embedded systems, computer hardware. This will ensure that both of us will profit from the supervision: You get technical guidance and I get support for my research or an opportunity to peek into industrial problems in case of an external final thesis/internship.
Why are industrial contacts important for me? My goal as a researcher is to create knowledge for the society and train next generation of researchers. To hire a researcher or buy large research equipment I typically need to write project proposals to organizations that distribute money to good research ideas. These organizations often prefer academic industrial collaborations, because industry has typically the infrastructure to make a technology available for the society in a sustainable way. This is one of the reasons why I still encourage MSc students working with a interesting company to ask for supervision.
Sometimes you may not be able to find a technical supervisor at DTU in the area you are doing your internship of final project, e.g., because the person does not have any capacity. Then, I can still help you by supporting the organizational elements of your internship and project, e.g., registering, general advice, writing the report.
I recommend to prioritize a supervisor who has technical expertise in the domain you are working. I once saw that a student wanted to choose a kind and understanding supervisor, even the supervisor lacked technical expertise. Even this is important, I would still recommend finding an supervisor who can give you valuable technical feedback in your area. You could probably look over the bad manners of the supervisor in the limited time period. Feedback and assessment are the fundaments of education, after all.
When cold contacting professors, attach a concise and well-written project proposal. This could improve your chances of being accepted.
Submissions#
Final thesis submissions are done through https://student.dtu.dk
Grading#
BEng internship grading is only based on the submitted documents. For final projects:
based on the oral defense + report or thesis
max 10 days after the submission
with external censor
7-grade scale with external censor
-
if group work:
first all group (you have the right to opt out)
then individual
-
For the Master’s thesis:
based mainly on the academic content
formulation and spelling also count, but less than the academic content
Collaboration/feedback infrastructure for the report & project#
Use a code repository from the beginning and give access to me. This eases feedback.
If you like LaTeX, I recommend using typst for your thesis and proposal.
-
Use a document writing environment that eases commenting and feedback, e.g.,:
git repository (with pull requests)
a shared Word document, e.g., Office365
Overleaf
Supervision style#
Depends on whether the topic is in my area or not.
Topic is not in my area of research or expertise#
As a BEng student this will be the case in most situations. I will only contact you to visit you roughly one month after your project has started. You will be mainly supervised by your supervisor at the company. I will mainly help you with your report.
I prefer mails over meetings. If you still need a meeting, you can book me.
Topic is in my area of research#
In the first weeks, we will have weekly meetings, then we can adjust the frequency. I reserve about 15 to 30 min for each student every week.
During our meetings you can expect questions from me, because I like to test the understanding of my students.
Final-project and internship proposal#
If you want to work with me, then please send a project proposal containing:
-
Background
Where does the problem come from?
-
Problem
What do you want to solve? Is the problem already solved? Why should this problem solved (again)?
-
Goals:
if research: Which research questions do you want to answer?
if engineering: What should your product/solution achieve?
-
Methods:
Which methods will you use to achieve your goal? Example: I will implement a C++-based solution. Its performance will be tested using a test environment
-
Tasks:
Which tasks do you expect on your path?
What is your (weekly to monthly) schedule?
-
Resources/tools:
Which resources/tools will you use to achieve your goal.
Registration#
DTU BEng students#
If we agreed to work together, then use the corresponding form for registration:
After filling them out, send them to Trine.
DTU MSc students#
We have to do it manually, so ask me.
Internship for external students#
You are welcome 🙂. You have to find a funding for your visit, e.g., Erasmus+. Our department needs up to three weeks to process our application, so contact me weeks before your deadline. I can propose a topic for you.
Other points to pay attention to#
Pay attention that DTU employees must take about three weeks vacation in July. So, do not plan your examination/internship in July.
Recommendations for the written work#
I collected below my recommendations from my feedbacks:
Note
Most of these are recommendations. If you disagree and do it your own way, that is fine! Just let me know where you have different taste/opinions.
-
Use a cloud-based editor / a git repository for your work instead of sending a PDF each time for feedback. Advantages:
I have the latest version.
Cloud-based solutions tend to have a feedback/annotation functionality that can be tracked.
Avoid many empty/non-actual-content (e.g., content, acknowledgements, list of figures) pages before your table of contents and the content. Most of the readers are interested at the actual content — table of contents and the content from there. So reduce the number of empty/non-actual-content pages or move them to the end of the document.
Your matriculation number should be clearly visible in the proximity of your name in the first pages.
Do not write the names of your supervisors larger than your name. This is your work.
Abstract must introduce the problem and contain key results.
Introduction should be written in a language that can be understood by someone with an academic education. In other words it should not assume special knowledge.
If you include theory, then only write about which is relevant to understand your work. Some students (including me back then 🙂) fill many pages with theory, because it can alleviate the anxiety of not having enough content. You should focus on your own contribution.
If writing in the industry, you may feel the urge to describe the company as well including its history. Only include relevant details about the company, e.g., how does it relate to the company, and move the rest to the appendix.
-
After reading the introduction the reader should:
know where the problem comes from
find sentences about what to expect in each chapter
2-4 pages are sufficient!
Literature research about existing related work should be included.
Make use of diagrams! Diagrams give an overview of the components of your work and the relations between them in a compact way.
Do not include long pieces of code. It disturbs the reading flow. If you need to show a concept using your code, include only the relevant few lines.
If you want to publish your work later, pay attention that (1) you attribute the visuals (2) use visuals that have a license that allow publishing, e.g., Creative Commons.
Use a language model only after you have written your text to improve your writing. See the process of writing your report as part of your education.
You must cite your references inline and citations should be clickable. Ideally, the reader should be able to access your reference by clicking a link right after your claim. The goal is to make fact-checking by the reader as convenient as possible.
-
If you use formulas:
present the formula
describe each formula component.