Frequently Asked Questions (FAQ)

Please note that both FYP and UROP also have specific FAQs within their documentation.  Please access the secondary menu (usually on the right hand side for desktop browsers) to learn about specific, frequently asked questions for those project courses.

How do I get good reference letters?

Good references (and their eventual letters of recommendation) are an important part of getting that first job or securing a strong graduate school placement. Project courses provide a key avenue for this important outcome in your CV and résumé, as you get to work on one with faculty members and their student teams. Devoting sufficient time, energy and initiative towards your project courses greatly enhances your ability to achieve a strong outcome in your project, and the ability of your faculty mentor to write a solid recommendation on your behalf.

What do I need to get into graduate school?

Strong graduate schools (inclusive of NUS) generally have two separate criteria for admissions. There is a minimum standard for consideration, usually fanthomed from standardized test scores and undergraduate transcripts. After the consideration bar, staff or faculty individually assess applicants as to their fit for the school. This step is subjective and admissions faculty and staff do not just look at academic credentials but also examine the prospective candidates fit for the school, in terms of diversity and non-traditional credentials. The acceptance bar for graduate studies can be influenced by letters of recommendation from coursework and project mentors, as well as internship advisors. Hence, doing well in project courses is a good pathway towards graduate admissions. Publishing original work through a thesis or peer reviewed publication are also strong quality markers that world class doctoral programmes will consider.

How do I balance research with coursework?

Project and time management is a key factor in project course success. Unlike coursework, project courses do not necessarily have clearly demarcated deadlines and deliverables – each project is a unique mix of challenges, prior work and personnel. As such, coursework that has a much more regimented timeframe tends to be prioritized by students given their concrete deadlines. This means that much a project course progress may be best loaded to the beginning parts of the semester when deliverables for coursework are significantly less. Make sure to balance sufficient time for reporting and communicating results in a polished manner, as the communicated results are what evaluators judge you on and often not the effort made.

What is the role of a supervisor?
The faculty supervisor serves to give you the framework to do your project.That said, each student-faculty pairing is unique and when you look for a project supervisor, consider interviewing multiple prospective faculty members before deciding. Each has a different style of supervision and a good fit will enhance your performance. For FYP and UROP, you should “shop around” during the discussion period.
What about CS2309, CP3106, CP3108 and R courses?
See the project course overview page and the respective NUS bulletin pages.
Is research mostly non-coding?
Yes. SImilar to most IT projects, coding is a small fraction of the total time on a project. Theoretical projects may not involve coding whatsoever. Certain areas of CS/IS/BZA have more exposure to systems work and require more development, and others less. Remember, research involves the entire research lifecycle: from problem inception, to investgation, to innovation, and finally, to communication. Coding is only a part of the second and third steps.
How can someone with zero prior research experience contribute to complex research projects?
Most complex research projects involve multiple, interdependent parts. Your faculty mentor is responsible for guiding your involvement in scoping your involvement such that you get to experience part or all of the research lifecycle. Significant contribution also usually requires proportional background knowledge in the particular research area. A key point to note is that modular coursework often is too broad to give the necessary background for research. To prepare for a particular project, you might ask faculty to prescribe specific pertinent readings rather than take a background course.
What are the pros and cons of doing research pathway as opposed to internship pathways for SoC students?

At its core, the research pathway embues students with the capability to do independent study: that is to take vague interests or challenges, formalize them into crisp problem statements and conduct inquiry to solve the problems. Research is also specific to subdisciplines of computing, and hence students will develope expertise in their area of exploration. These outcomes are important for jobs requiring independent initiative and skills development.

Internships complement the research pathway by allowing students to acquire particular in-demand skills and building their commercial exposure to real-world problems and career opportunities. This pathway can help students find a better match for the first career and often higher pay.

What all should students be doing in their undergraduate studies to pursue a career as a research scientist in companies?
Both the research pathway of research-oriented project courses as well as research-oriented interships will help develop your capability to pursue an eventual research scientist position. Many companies will require evidence of your ability to conduct independent investigations, and hence higher studies through doctoral or masters are often required.
Do I need to have done UROP or FYP to go to graduate research?
No, generally you do not need to UROP or FYP to go to graduate doctoral programmes or masters by research. While some strong schools like NUS have research opportunities for undergraduates, others may not have such opportunities. UROP and FYP can enhane your chance of admission and distinguish yourself from other applicants with similar backgrounds and academic performance.
Hypothetically, which one would you pick: a research area with a beautiful goal but messy process/method, or one with a really elegant process/method but whose goal you are unsure of?

This is a personal preference, and this tradeoff is evident in certain areas of computing as well. While not exactly an answer to this question, often prodcution computing systems need to use heuristic tricks and optimization beyond core, elegant algorithm or theory in order to work at scale or within time-bounds.

We welcome additional questions to add (of course with their answers) to this FAQ. Please send your queries to the Undergraduate Office (