← Back to blog

A Fucked 2025/2026 Recruiting Season

February 19, 2026

2025 had been the most grueling recruiting season in my life. I had probably done 30+ rounds of interviews, and faced dozens of rejections despite passing multiple rounds. This blog will consist of my thoughts, and be a reflection on how these went. The ups and downs, and what events occurred for me to receive almost no interviews in 2024, to getting dozens in 2025.

The Beginning

Reflecting on what changed from 2024 to 2025 for me to receive interviews was projects with users and attention. Prior to September 2024, I never really had a 'proper' full-stack project because AI-assisted coding wasn't as popularized as now. But ever since I began using Cursor, I was finally able to populate my resume with projects that actually worked. And this allowed me to get my first software engineering job at Caivan, as well as receiving offers from other smaller companies which I rejected.

Early 2025 was when I finally started getting attention from companies. At Carleton Hacks, Rodney, Alvina, and I created the Rizz Glasses. We then posted this on LinkedIn, which received 100,000+ impressions within the first month. As a result of that, I began receiving coffee chat requests, and interviews from a few smaller companies. These included:

This was the first time I had ever talked to startups, and had no clue about funding, the qualifications required, and the type of people they are looking for. They ultimately want builders and people who are obsessed with customers, but at the time I didn't emphasize these enough, which may have resulted in these rejections.

In the meantime, I had also begun doing an extensive amount of LeetCode. During the winter term, I had done around 250 problems, finally being able to comfortably solve mediums without assistance. Unfortunately for the next several interviews, I didn't receive a single LeetCode question.

First Shopify Interview

By the end of April, my internship at Caivan had ended and I received my coding interview with Shopify... which resulted in:

  • Shopify - Rejected after coding interview

This rejection stung, mainly because I solved the question correctly. But I may have not used enough AI, and probably because I couldn't hear the interviewer well enough since he had a bad microphone. Regardless, I should have spoken up.

Summerloo & YC Startup School

Summerloo had started and it started a little rough. In cycle 1, I had 1 interview with Super.com, and most of my friends had already received their Shopify offer, which meant they didn't have to go through WaterlooWorks.

  • Super.com - Rejected after coding interview

Things changed however, when I went to the first-ever YC Startup School. I spent the entire time there socializing. I was invited to a YC founder dinner with Eric Levine, went to a YC founder's house, and heard how many successful people navigated their careers. This advice ultimately changed the way I presented myself to companies, which resulted in:

Again, I still haven't fully understood the way to pass all my behavioural rounds with these startups. I have been able to explain my projects very well, but didn't glaze hard enough about their product and vision.

When cycle 2 rolled around, I received 5 interviews. And they went quite well. I won't list all of them, but I received an offer from Compass Digital. A subsidiary AI lab for a public company Compass Group. They have a really good team, and it seemed like a great learning experience. + it was in Toronto.

TopClip & Soverin

From June to August, I didn't bother interviewing for other companies and focused exclusively on school work. Until a company I cold DM'ed in March, Soverin, sent me a message to do a work trial which I accepted. It lasted around a week and a half, and they paid me $8,000 to build TopClip.ai. I accepted, and was convinced to join their team as a founding engineer and reneged my offer at Compass Digital.

This might've sounded stupid, but there was a rationale behind this decision of reneging a well-paying Toronto-based job. Within the first month of releasing TopClip, it had gotten 30,000+ users, then 100,000+ after 4 months. Being able to solo build something used by this many people is unheard of at larger companies, hence why I made this decision.

Founding Engineer Interviews

Around this time, I was supposed to be in New York working with Soverin, but due to 'budget constraints', they made the decision to not fund my visa, nor did they raise a seed round in time for me to fully commit to the company. I was also contacted by several headhunters, and I had begun receiving multiple interviews for founding engineering roles with the possibility of taking a gap year if it was worth it.

This is around the time when I knew exactly how to present myself and pass almost all screening calls. I would present building TopClip.ai to 100k users, which impressed almost all founders and recruiters. It signified autonomy and the ability to build sophisticated apps for a large number of users.

Hence, I got interviews for:

  • Hanover Park - Rejected after 3rd round technical
  • Anvara - Rejected after 4th round technical

Hanover Park had a really strong team, an ex Goldman Sachs co-founder, an exited co-founder, and raising their series B. They asked a medium BFS graph problem which I unfortunately wasn't able to solve on time.

Dream Companies

The biggest upset of the recruiting season was fumbling 3 companies that were on the top of my list. Mainly because I had used their products in the past, they had cracked teams, and were ones I would work for post-graduation. These were honestly dream companies I wanted since first year.

  • Vercel - Headcounted after 2nd round technical
  • Slash (YC S21) - Rejected after 2nd round technical
  • Whop - Rejected after 2nd round technical

Vercel was just unlucky, but Slash and Whop had genuinely difficult interviews. They asked implementation questions that would have taken me hours to do. Slash asked a topological sort implementation of cascade delete, and Whop asked a question on matching receipts, users, and transactions from three streams of data.

Work Trials

Around this time, I had also done initial interviews with 2 companies, and completed work trials. These all would have been offers, but I decided not to take them because I found it more worth to do 3/4 more 4 month internships instead.

Winter 2026

When CoreThink ended, I came back to Waterloo around late January 2026. WaterlooWorks cycle 1 had already started... which went a bit rough:

  • TextQL - Cancelled internship hiring after 2nd round takehome
  • DataCurve - Too lazy to do 4 hour takehome
  • HeyGen - Rejected after 1st round technical
  • SafetyKit - Rejected after 3rd round 4hr technical
  • Faire - Rejected after 1st round technical
  • Shopify - Rejected after life story

For the first time I had received cycle 1 interviews for great companies. All the companies were post Series A and had solid teams. What sucked even more is that I finally passed the Shopify technical interview because I knew what to do, but failed the behavioural. Tbh, I probably sounded like a tech nerd, which might not have been what they were looking for.

Current

As I write this, it's cycle 2.

The stealth startup has a very solid team based in NYC, consisting of ex-Scale, Google, Palantir, Goldman Sachs, and Databricks members. Though, I would prefer working in SF this summer since all of my friends are there, and a couple YC founders who DMed me to meet up.

Next is a technical and possible work trial. But we will see what comes next.

To be continued...