why i left ETH Zurich (and why it was the best decision i've made)

3am. staring at a proof by induction for a discrete math problem set. my eyes are blurring. i've been at this for four hours.

meanwhile, in my notes app: twelve project ideas collecting dust. a half-finished SaaS. a CLI tool i sketched out on a train ride. a side project that three people asked me to build.

all gathering cobwebs because i'm proving that some function is O(n log n) for the third time this semester.

something wasn't adding up.

the prestige trap

ETH Zurich is one of the best universities in the world. top 10 globally. einstein went there. it produces world-class researchers, engineers, and scientists. getting in is hard. staying in is harder. the dropout rate in CS is... let's just say "significant."

and i was doing fine. not crushing it, but passing. keeping up. doing the work.

but i wasn't building anything.

ETH's CS program is deeply theoretical. rigorous mathematical foundations, formal methods, algorithm proofs, abstract algebra. this is intentional — they're training researchers and foundational thinkers. if you want to push the boundaries of computer science itself, ETH is one of the best places on earth.

the problem was: that's not what i wanted to do.

i wanted to build things. ship products. work with real users. break production and fix it at 2am. deploy on friday and deal with the consequences. learn by doing, not by proving.

ETH was the right school. just not for me.

the decision everyone questioned

i transferred to FHNW — the University of Applied Sciences and Arts Northwestern Switzerland. a part-time program that lets you work alongside your studies.

the reactions were... predictable.

"you're leaving ETH?" "are you sure?" "but the ranking..." "you'll regret this." "ETH looks so much better on a CV."

i heard all of it. from friends, family, acquaintances. some people said it to my face. others said it behind my back (which, honestly, i prefer — less awkward).

the concern was genuine. from the outside, it looks like a downgrade. ETH is globally ranked. FHNW is... not. nobody outside Switzerland (and honestly, plenty of people inside Switzerland) has heard of it.

but rankings measure research output, not whether a program is right for a specific person. they tell you about the institution, not about your fit with the institution.

what happened after

here's what i shipped in the time i would have spent on ETH problem sets:

joined Migros as a cyber defense engineer. the largest retailer in Switzerland. real production systems, real threats, real incident response. the part-time program at FHNW made this possible — i work and study simultaneously. at ETH, this wasn't an option.

built and shipped TTStats. a platform tracking 5,500+ Swiss table tennis players and 69,000+ matches. 1,200+ monthly active users. grew organically through word of mouth at tournaments. won 2nd place at a startup award. went from web app to native iOS and Android apps with Kotlin Multiplatform.

built Bunzlimeter. a fun webapp that went semi-viral — 10,000+ users, got featured on SRF radio (Swiss national broadcaster). the kind of thing you build in a weekend when you actually have weekends.

released 3 audio plugins. because apparently i also make music production tools now.

rebuilt my portfolio from scratch. the site you're reading this on.

contributed to open source. including finding and responsibly disclosing a security vulnerability in Uptime Kuma (86k GitHub stars) that led to a published CVE.

maintained top grades at FHNW. because when you're studying something applied and building things simultaneously, the coursework makes sense in a way pure theory never did for me.

none of this would have happened at ETH. not because ETH is bad — because i would have been too exhausted from problem sets to build any of it.

the prestige of your university doesn't matter if you're too burned out to build anything

this is the line that tends to resonate with people.

a degree from ETH opens doors. absolutely. the brand carries weight. employers know it. recruiters filter for it. this is real and i don't want to pretend otherwise.

but a degree is one signal among many. and if getting that degree costs you every evening and weekend for four years — time you could spend building, shipping, learning by doing, developing a portfolio, growing a network — you need to ask whether the signal is worth the opportunity cost.

for some people, the answer is absolutely yes. ETH produces incredible engineers and researchers. the people who thrive there are the ones who love the rigor, who get excited by proofs, who want to do PhD-level thinking.

i'm not that person. and admitting that was the hard part.

sometimes the "downgrade" is the upgrade

switching to FHNW gave me three things ETH never could:

time. time to build side projects. time to work a real job. time to contribute to open source. time to exist as a person and not just a student.

applied context. every concept i learn in class, i can immediately apply at work. networking theory becomes real when you're configuring firewalls. security concepts become real when you're responding to incidents. the feedback loop between theory and practice is instant.

proof of ability that isn't a diploma. when i apply for jobs, i don't lead with my university. i lead with TTStats (1,200+ users, startup award), Bunzlimeter (10k users, radio feature), open-source contributions (CVE in an 86k-star project), and my work experience at Migros. the portfolio speaks louder than the institution name.

the honest caveats

i'm not saying "drop out of ETH." i'm saying "be honest about what you want and what you need."

ETH is genuinely world-class. if you want to do research, go to grad school, work on foundational CS problems, or join a company that specifically filters for elite-university graduates — stay. you're in the right place.

but if you're sitting at 3am staring at a proof and your notes app is full of unbuilt projects... maybe the problem isn't discipline. maybe the problem is fit.

the hardest part of leaving wasn't the logistics. it was letting go of the identity. "ETH student" carries social weight in Switzerland. giving that up felt like admitting failure.

it wasn't failure. it was alignment.

the aftermath

i posted about this decision on LinkedIn. it got 14,413 impressions — my most viral post ever. hundreds of reactions. dozens of DMs from other students saying "i needed to hear this."

turns out a lot of people are sitting in programs that look great on paper but don't match who they are. most of them just need permission to consider the alternative.

here's your permission: the "downgrade" might be the upgrade.

---

sometimes the right school isn't the best school. it's the one that lets you be who you actually are.