🚀 Before starting
🔍 Finding a project
The first step in writing a thesis is to find a project to work on. The four most common ways are:
- Find a project at a company, maybe where you are working as a student assistant
- Find a project through https://dtu.jobteaser.com
- Propose a project yourself
- Contact a professor or researcher at DTU for project ideas
Regardless of what way you find a project, it is important that you find a project that you are interested in! You will be working on the project for several months and burnout is a real thing. So, even if you are not in charge of formulating the initial project idea, you can still influence the project during the process to make it more interesting for you. Remember, in the end it is your thesis 🎓
When considering project, think about what parts of computer science that you find most interesting. Very broadly, a thesis within computer science can be divided into four categories(1)
Yes, this is a brute simplification of the field of computer science, but it is a good way to
think about what you find interesting and in my experience is still fairly covering.
-
Theory: This is the mathematical and theoretical part of computer science. It is about understanding the fundamental limits of computation and algorithms. If you like math and theory, this is the place for you.
-
Software: This is about understanding and working with software. It is about understanding how to design, build and maintain software. If you like programming and software development, this is the place for you.
-
Data: This is about understanding and working with data. It is about understanding how to store, process and analyze data. If you like working with data, this is the place for you.
-
Modelling: This is about understanding and working with complex systems. It is about understanding how to model and simulate systems. If you like understanding complex systems, this is the place for you.
You will most likely be working with a combination of these four categories, but it is a good idea to think about what you find most interesting. For example, if you are doing a project in the field of machine learning, you will most likely be working with data and modelling as the primary focus. That said, you can put more focus on software by implementing not just a model but a hole ML system and in the same way you can put more focus on theory by understanding the theoretical limits of the model you are using.
🧙♂️ Finding a supervisor
Regardless of how you find a project, you need a supervisor from DTU to register the project and guide you through the project (yes, also for company projects). The DTU rules state that:
Main supervisor
The main supervisor for a thesis must:
-
Be an employee at DTU as a VIP with research obligations and one of the following job categories: assistant professor, associate professor, researcher, senior researcher, senior adviser, or professor,
-
Hold a permanent position (with the exception of assistant professors and researchers)
-
Have completed DTU's course 'Supervision of larger projects at DTU' or similar pedagogical competence development at the university level e.g. a course in PhD supervision; for assistant professors, however, it is sufficient that one is in the process of UDTU, and
-
Have in-depth knowledge of the study conditions at DTU.
So not all employees at DTU can be your main-supervisor. That being said, if you find a supervisor that does not meet the above criteria (PhD student, Postdoc or tenure-track assistant professor) it is possible that they can still supervise you, as long as you find a college that meets the above criteria to be your main-supervisor on paper.
If you are doing a company project, you will also need a co-supervisor from the company. The co-supervisor does not officially hold any responsibilities, but they are very much expected to help you with the practicalities of the project and provide you with the necessary data and information.
You can in general find supervisors by looking at the page belonging to one of the 10 sections belonging to DTU Compute:
| Section | Focus | Link |
|---|---|---|
| Algorithms, Logic and Graphs | Algorithms, logic, graph theory, and discrete mathematics | https://www.compute.dtu.dk/sections/algolog |
| Cognitive Systems | Human–machine interaction, signal processing, machine learning | https://www.compute.dtu.dk/sections/cogsys |
| Cybersecurity Engineering | Secure systems, cryptography, networks, and cyber-physical security | https://www.compute.dtu.dk/sections/cybsec |
| Dynamical Systems | Modeling, forecasting, optimization, and control of dynamic systems | https://www.compute.dtu.dk/sections/dynsys |
| Embedded Systems Engineering | Hardware–software co-design, edge/cloud systems, dependable computing | https://www.compute.dtu.dk/sections/emsys |
| Mathematics | Pure and applied mathematics, algebra, analysis, geometry | https://www.compute.dtu.dk/sections/math |
| Scientific Computing | Numerical algorithms, modeling, simulation, and high-performance comp. | https://www.compute.dtu.dk/sections/scicomp |
| Software Systems Engineering | Software engineering, programming languages, formal methods | https://www.compute.dtu.dk/sections/sofsys |
| Statistics and Data Analysis | Statistics, bioinformatics, data analysis, pattern recognition | https://www.compute.dtu.dk/sections/stat |
| Visual Computing | Image analysis, computer vision, graphics, 3D modeling | https://www.compute.dtu.dk/sections/viscom |
🤝 How to approach a potential supervisor
Researchers are busy people and in many cases they already have too many students to supervise. Therefore, it is important that you approach potential supervisors in the right way. When contacting a potential supervisor, always do it from your DTU student email address, else you risk being send to the shadow realm (spam folder). Feel free to include a CV and or academic transcript, it is not required, but it some supervisors will ask for it.
Here are some common scenarios you might find yourself in when contacting a potential supervisor.
Case 0: How NOT to approach a supervisor
Hi, my name is X, I study Y and I am looking for a thesis project. Do you have any projects available?
This kind of email is very likely to be ignored. Many students are searching for thesis projects, and unless the researcher has a current list of available projects posted online, you probably won’t get a response. Don’t take it personally—researchers receive a huge number of these emails and simply don’t have time to answer them all. In fact, replying often means starting a longer conversation, which can be even more time-consuming.
Case 1: You have a project idea
Dear Professor Z,
My name is X, and I am a Y student at DTU. I have a strong interest in [specific area of research], and I recently came up with an idea for a thesis project that I believe aligns well with your expertise in [related field]. The project involves [brief description of the project idea].
I would greatly appreciate the opportunity to discuss this idea with you further and explore the possibility of you supervising my thesis. I am eager to contribute to your research group and learn from your experience.
Thank you for considering my request. I look forward to hearing from you.
Best regards, X
Check the research pages of relevant departments or research groups, as well as the individual pages of academic staff members (starting with lead researchers or professors). The goal is to identify one or more potential supervisors who work in your area of interest or a related research field. You can then contact these potential supervisors by email. The email should include:
- A brief introduction about yourself and your academic background
- A list of courses you have completed, along with your grades
- A 1-2 page description of your project idea: including the problem statement, its relevance, your proposed approach, and what makes your idea novel
Providing this information helps the potential supervisor understand both you and your project, and serves as a solid foundation for further discussion.
Case 2: You have a project idea in collaboration with a company
Dear Professor Z,
My name is X, and I am a Y student at DTU. I have a strong interest in [specific area of research], and I am currently working with [Company Name] on a project that I believe aligns well with your expertise in [related field]. The project involves [brief description of the project idea].
I would greatly appreciate the opportunity to discuss this idea with you further and explore the possibility of you supervising my thesis. I am eager to contribute to your research group and learn from your experience.
Thank you for considering my request. I look forward to hearing from you.
Best regards, X
Check the research pages of relevant departments or research groups, as well as the individual pages of academic staff members (starting with lead researchers or professors) to find someone who does research or has expertise related to your project. You can then contact potential supervisors by email. The email should include:
- A brief introduction about yourself and your academic background
- A list of courses you have completed, along with your grades
- A 1-2 page description of the project (including the problem statement, why the problem is relevant, your proposed approach, and what makes your idea novel), as well as details about the company and the name of your company supervisor(s)
Keep in mind that a project in collaboration with a company may require more time to define and officially register, as an agreement must be reached to meet both DTU's and the company's expectations.
Case 3: You don't have any project idea but you have a subject you like
Dear Professor Z,
My name is X, and I am a Y student at DTU. I have a strong interest in [specific area of research] and would like to explore potential thesis projects in this field. I am eager to contribute to your research group and learn from your experience.
I would greatly appreciate the opportunity to discuss possible project ideas with you and explore the possibility of you supervising my thesis.
Thank you for considering my request. I look forward to hearing from you.
Best regards, X
When searching for a supervisor, check the research pages of relevant departments or research groups, as well as the individual pages of academic staff members (starting with lead researchers or professors) to find someone who works in your area of interest. For example, if you are interested in a subject covered in a particular course, it makes sense to start by contacting the teacher(s) of that course. You can then contact potential supervisors by email. The email should include:
- A brief introduction about yourself and your academic background
- A list of courses you have completed, along with your grades
- A short description of the subject you are interested in
- The type of thesis project you would ideally prefer (theoretical, practical, or a combination)
Providing this information helps the potential supervisor understand your interests and background, and serves as a good starting point for further discussion.
Case 4: You don't have any project idea, but you have a subject you like
Dear Professor Z,
My name is X, and I am a Y student at DTU. I recently completed your course [Course Name], which I found particularly exciting and in which I performed well. I am very interested in [specific area of research related to the course] and would like to explore potential thesis projects in this field.
I would greatly appreciate the opportunity to discuss possible project ideas with you and explore the possibility of you supervising my thesis.
Thank you for considering my request. I look forward to hearing from you.
Best regards, X
If you are unsure whom to contact, consider reaching out to the teacher of a course you found most exciting or performed well in. Alternatively, you can contact your head of studies or the study coordinator for your program, as they may have suggestions for potential supervisors.
👓 Read the rules
Yes, it is super boring to read the rules, but it is important to know what you are getting into and you can often find questions to a lot of the questions you might have. So take your time to read them at least once. The rules may differ a bit from studyline to studyline, so make sure you read the rules for your studyline, which you can find here:
https://student.dtu.dk/en/programme-specifications
Especially, be aware of learning objectives for your thesis, as they are the basis for the assessment of your thesis. They will look something like this:
Learning objectives
A graduate of the MSc programme from DTU:
1️ can identify and reflect on technical scientific issues and understand the interaction between the various components that make up an issue
2️ can, based on a clear academic profile, apply elements of current research at an international level to develop ideas and solve problems
3️ masters technical scientific methodologies, theories, and tools, and can take a holistic view of and delimit a complex, open issue, put it into a broader academic and societal perspective, and, on this basis, propose a variety of * possible actions while considering sustainability
4️ can develop relevant models, systems, technologies, and processes aimed at solving technological problems
5️ can communicate and mediate research-based knowledge both orally and in writing
6️ is familiar with and can seek out leading international research within their specialist area.
7️ can work independently and reflect on own learning, academic development, and specialization
8️ masters technical problem-solving at a high level through cross-disciplinary teamwork, and can work with and manage all phases of a project – including preparation of timetables, design, solution, and documentation
Learning objectives
The BSc project from DTU must demonstrate that you:
1️ can work independently and is able to structure a major project, including meeting deadlines and organizing and planning the project work
2️ can summarize and interpret technical information and is fully familiar with technical problem solving through project work
3️ is able to work with all project phases, including the preparation of proposals, solutions, and documentation
4️ is able to independently acquire new knowledge and adopt a critical approach to the acquired knowledge and carry out relevant and critical information searches, and on this basis find the right methods to shed light on the problem in question
5️ is able to communicate technical information, theory, and results in written, visual/graphic, and oral form
All these learning objectives are fairly meta, but they can interpret as you need to master every aspect of doing a thesis from the technical aspects to the communication and project management.
In addition to the general learning objectives, each study line has its own programme specific learning objectives. These are much more concrete and describe the technical competences a graduate from that specific study line should have. Your thesis does not need to cover all of them, but it is a good idea to check that your project fits within them. As an example, here are the programme specific learning objectives for the MSc in Human-Centered Artificial Intelligence:
Programme specific learning objectives (MSc in Human-Centered Artificial Intelligence)
A Master of Science in Engineering (Human-Centered Artificial Intelligence):
1️ can delimit a complex open issue and select technological solution options in relation to data, software, interaction and computer graphics
2️ can conceptualize, design, implement and validate software solutions
3️ can model and design interaction between humans and computer systems
4️ can apply elements of artificial intelligence and machine learning, including generative artificial intelligence
5️ can apply visual computing in the development of digital media products
6️ can combine technological and commercial knowledge for the design of digital media products
7️ can incorporate elements of new research, and assess ethical, societal, and responsible artificial intelligence implications, in the development of innovative digital media products
8️ can model user needs hierarchically as outcomes, activites and prioritized tasks and apply lean methods to iteratively validate solutions based on rapid prototyping
9️ can develop business models to optimize value in digital media products
10 can formulate quantitative models and use them to solve technological issues in relation to data, software, interaction and computer graphics
11 can apply a high-level programming language (e.g. Python) to utilize web APIs and model APIs for data acquisition and AI integration
12 can apply algorithms to recommend media (text, audio, video) according to user preferences and user context (friends, mood, location, etc).
You can find the programme specific learning objectives for the most common computer science related study lines at DTU here (the bachelor programme specifications are only available in Danish, where the section is called Retningsspecifikke mål for læringsudbytte):
| Level | Study line | Link |
|---|---|---|
| MSc | Computer Science and Engineering | link |
| MSc | Human-Centered Artificial Intelligence | link |
| MSc | Mathematical Modelling and Computation | link |
| MSc | Business Analytics | link |
| MSc | Autonomous Systems | link |
| BSc | Software Technology (Softwareteknologi) | link |
| BSc | Artificial Intelligence and Data | link |
| BSc | Cybertechnology (Cyberteknologi) | link |
| BSc | Applied Mathematics (Anvendt matematik) | link |
| BEng | Software Technology (Diplom Softwareteknologi) | link |
If your study line is not in the list, find it through https://student.dtu.dk/en/programme-specifications.
💾 Registering the project
The project needs to be officially registered by your supervisor to DTU study administration. In practice this should be done about two weeks before you start the work. For your supervisor to register the project they need some information.
Registration information
-
Title: The overall title for the project
-
Description: A short (5-10 lines) description of the project, that motivates the project and gives an overview of the work to be done. You do not fix the project description in stone, but it is a good idea to have a clear idea of what you want to do. The description should also make it clear that the project falls within the programme specification of your study line, i.e. that it matches the programme specific learning objectives (see Read the rules).
-
Number of ECTS: A master thesis can be 30, 32.5 or 35 ECTS (the default is 30 ECTS). A bachelor thesis can be 15, 17.5 or 20 ECTS.
-
Start date: The date you start working on the project. If you start in the spring, you are expected to start either in the first week of january or if you are having a 3 weeks course, then the first week after 3 weeks period have ended. Similarly, if you start in the fall, you are expected to start in the first week of August or the end of August depending on if you have a 3 weeks course in June. That said, you can register on alternative days but you need to provide a reason for it.
-
If company project please also provide:
- Company name
- Company address
- Company contact person
- Company contact email
- Company contact phone number
-
Student information: Your student number and if it is a group project, the student number of your group members.
-
Supervisor information: If you have more than one supervisor, you need to provide the email-initials of the ones that are not registering the thesis.
Note that the title and description should be provided in both Danish and English.
🗣️ Setting expectations
As the final step before starting the project, you should set expectations with your supervisor. This is important to make sure that you are on the same page and that you know what is expected of you. Consider the following questions:
-
What is the expected outcome of the project?
-
What grade do you expect to get?
-
What should the meeting schedule be? Regular vs. ad-hoc meetings? Physical vs. online meetings?
-
What means of communication should be used? Email, Teams, Slack, etc.?