Reflecting on applications and interviews for Summer 2025

As a product of my growing comfort managing undergraduate student projects, I am pursuing several projects over the summer. This meant that, in the last few months, I reviewed a (in my mind) significant number of applications and interviewed quite a few students. For context, I reviewed applications for:

• An ROP project (with 3 sub-projects) for the 2025-2026 academic term
• 3 different CSC494H1 projects for the Summer 2025 term
• 1 work study position for the Summer 2025 term

In this post, I reflect on my decision making process. While you may consider this an “advice” post, there is the caveat that these are my opinions and this is a reflection. Based on some LinkedIn responses, I certainly have a different experience and seem to value different things. Still, I think sharing this reflection is useful.

Application #

All my postings required the applicant to provide a resume, cover letter, and academic transcript. So I’ve divided this section into a reflection on each of those documents. But one issue I encountered with work study applications was that some were incomplete! For example, the student would (I assume accidentally) upload a resume twice (once as the cover letter and once as the resume). Perhaps some obvious advice is to review the documents you submitted to make sure they are the correct files!

Resume #

I found that I did not read resumes. I did glance at them and, at some points, referred back to them looking for specific information. For example, when I wanted to align the skills of the applicant with what was required by the job.

When a resume was nicely organized, this was a breeze and gave me a positive impression on the candidate. I think there are a lot of resume templates out there, and I highly recommend using one (here’s one recommended by our DCS Ambassadors). Look for the templates that allow you to neatly organize your skills and experiences.

One definite plus to some resumes was the built-in links to other profiles, like LinkedIn or GitHub. In the case of LinkedIn, it’s nice to put a face to a name (especially if I’ve taught you before and the name looks familiar). If you have a LinkedIn with no profile picture, I recommend adding one. In the case of GitHub, it’s nice to get a quick glance at a candidate’s repositories. These were typically course projects, but sometimes they gave insights into other interests the student had.

I didn’t have much time to look through every applicant’s GitHub and/or LinkedIn profiles. But when shortlisting applicants to interview, those links could make a difference. For example, GitHub showed me one candidate was clearly an active programmer outside of courses. This helped confirm what they stated in their cover letter and resulted in an interview.

Cover letter #

I spent the most time for all applicants reading through their cover letters. In my mind, this was where I could get a true sense of the applicant. The best cover letters contextualized how their experiences (listed on their resume) actually connected with the job posting.

It was very obvious when a cover letter was simply copy-pasted, likely from an application (or many applications) to another job. These cover letters tended to focus on talking more about the candidate’s experience. But there were no connections to the job posting itself. In some cases, the cover letter was mostly generic, but there was some effort in connecting it to the job posting. Unfortunately, those connections tended to be superficial.

Overall, I did not interview candidates with these types of cover letters. It acted as an early filter. I highly recommend taking the time to reflect on why you’re a good fit for the job posted. And, if you know who’s posting the job, start with “Dear Mario” (or similar), not “To whom it may concern” (or similar).

Transcript #

Even though I asked for a transcript, this largely didn’t factor into most of my decision making. I say most because sometimes it did matter, but it very much depended on the posting. For example, if the job required a solid knowledge of fundamental CS concepts, then a decent grade in those courses is likely a requirement.

I never considered every grade in the transcript. For one, I don’t know every course at the University of Toronto. But also, that’s a lot of data points!

Depending on the job, I’d look at a handful of related grades to give me a sense of a students academic ability in a certain sub-field of computer science. But, at the end of the day, it’s difficult to use even a handful of grades as an indicator of mastery. So, really, I think I mostly used it as a red-flag marker.

Interview #

I typically structured my interview in three parts:

1. Introductions
2. Questions I ask
3. Questions the interviewee asks

I used this structure even when I knew the student, which is a little funny since I’m still using my canned introduction. But the best student introductions I heard went beyond the “I’m in X year pursuing Y Major and Z Minor”. These students took the opportunity to demonstrate other things, like their passion for the job or highlighting their relevant prior experience. This sometimes backfired (that may be too strong a word), in the sense that the student talked for too long and began answering questions I hadn’t asked yet. But the enthusiasm tended to leave a good impression on me.

I found students with the most experience tended to have the longest answers. Here, experience relates to other opportunities they pursued (like internships or clubs) and not which year they are in. Interviewing really is a skill - even if you’re not getting the job, simply the practice involved in going through an interview is very valuable. I could tell who’d interviewed before and who never had.

If you’re not landing interviews, practice with friends! Generative AI could come up with the questions, just feed it the job posting. This mock simulation, I think, is very valuable.

It was also clear who came prepared and who didn’t. I think there is almost always an opportunity to ask the interviewer questions. This is your chance to demonstrate not only your preparedness, but how you think and what you value. For example, a somewhat common question I received was whether the work was independent. These students valued working with others, which is one trait I wouldn’t have known unless they asked their question.

To sum up, I think there are quite a few ways to prepare for an interview. And, in a world of AI, there are tools that never existed before to help you. And yes, AI helped me craft at least an initial draft of my own interview questions (sometimes).

“Warm” emails #

I received quite a few emails regarding my job and project postings. I call these “warm” (as opposed to cold) emails because they’re not out of the blue; I am clearly looking for applicants!

Typically, the emails included an expression of interest from the student and attachments, like resumes and transcripts. Sometimes, the student wanted an opportunity to meet and discuss the posting. I suppose the question is whether these emails work? Did they move the needle?

I wouldn’t say there was a causal relationship between the email and the interview. Many times, the emails were sent well before the closing date, and I don’t seriously review applications until the posting has closed (so that I have all available applicants). My typical response to these emails was along the lines of, “Thank you for your interest. Please apply through the official channel.” By the time I got to reviewing the application, it was likely I had forgotten the names of the warm emailers.

But let me share an email (anonymised and with permission) that I think did a great job of capturing my attention,

Dear Professor Badr,

I recently submitted my application for some of your CSC494 research projects. I really resonated with how visual methods helped me study more effectively, and I think that I have the experience to help other students learn in the same way.

I’d be interested to hear more about your vision for the project. Some questions I have:

• Given that the project is entirely client-side, what are the logistics for instructors to create and save example visualizations? Would they work with configuration files like JSON?
• Is this envisioned as more of a sandbox where students can freely experiment with OS concepts? For example, with an MLFQ scheduling visualization, could students adjust parameters like number of queues, scheduling policies, and process characteristics to see how the system responds in real-time?

I’m excited about the possibility of contributing to this project and would appreciate any insights you could share. Thank you for your time and consideration.

The email begins with letting me know that the student has already submitted an application. This is great, because now I don’t have to double-check and ask the student to submit through the official channel. In addition, the email includes concrete questions about the project for me to answer. I think this did two things. One, it showed me that the student is not only interested in the project, but has already done some thinking about how they would pursue it.

Second, and perhaps subconsciously, I wanted to answer the questions. In contrast to emails that simply asked to meet and discuss, I think this made a world of difference. The discussion was already happening in their introductory email. All I had to do was continue it.

Conclusion #

This reflection helped me identify the things I value in an application (at least for now). But it also helped me realize that creating this many opportunities, reviewing applications, interviewing candidates, and finally making offers is a lot of work! I think it’s important for students to keep that in mind both in the materials they submit (keep them organized and relevant) and in the emails they send. There is, unfortunately, not enough time in a work day for me to respond to every email at the level of detail that I would like. Nor is there enough time to interview every applicant, and I wish there was a clear methodology for selecting the best applicants just by glancing at an application (there isn’t).

On the flip side, the market for opportunities is incredibly competitive. I try to remember that student’s have only so much time, themselves, to tailor their applications for each job posting. I know that some students opt for a “numbers” approach, sending as many applications out as possible. Still, after reflecting on my own process, I truly believe that identifying the jobs you really want to pursue, and tailoring your application for those jobs, is a better way forward.