Doing a Machine Learning PhD While Working in Japan

I recently completed my PhD in Computer Science at Tokyo Institute of Technology (now the Institute of Science Tokyo) as an American student on the MEXT Scholarship. Simultaneously, I continued my work as a software engineer and researcher at Google.

Because it’s relatively rare for an American student to complete their PhD program abroad—as compared to the number of international students that study in America—many people eventually asked me, “But why would you choose Japan?”

This comes with the implication that, given the American tertiary education system’s prestige, I was being foolish by not taking advantage of it. This point of view has merit. American schools do have incredible name recognition worldwide, as well as opportunities that are rare elsewhere. But I chose Japan, and would like to walk you through my experience to give you the pros and cons.

I admittedly did not have a typical background going into my PhD: I had a good fallback since I had already been working at a FAANG, I had meaningful savings, I already had research experience, etc. This made my time in Japan significantly easier and less stressful than for many of my peers. But it was still not a walk in the park.

Doing a PhD is difficult and isolating, which is exacerbated by being an international student in an unfamiliar country. And so, while I overall very much enjoyed my time there and have freely recommended it to many people, the decision to pursue a PhD in Japan should always be considered carefully.

Background

I’m an American and did my undergraduate degree in America, and before Japan, I was a PhD student in Korea. I didn’t complete my doctorate there, but left with a master’s to join Google full time after an internship.

My original research area was automata theory, but at Google I work primarily on on-device language models for keyboards. I have always inhabited an area between pure research and engineering and worked hard to maintain that balance.

I knew that I wanted to finish a PhD eventually, but I had a fairly specific constraint: if possible, I wanted to keep working part time while doing it. I knew there was some precedent for this arrangement at Google, but it was rare.

At the time I started looking seriously at doctorate programs again, I was working from my parents’ house during COVID-19. I had gone from Seoul, to New York for an internship, to Mountain View for a full-time role, and then back to my hometown in the American South. I had enjoyed some of those moves more than others, but the overall effect was that I had a pretty good sense of what I liked and disliked in a place to live. I loved dense cities, I liked being abroad, and I knew that if I was going to do another PhD, I wanted the location itself to be a meaningful part of the experience.

Why I considered Japan

When I started looking into PhD programs again, I was not particularly looking for something in Japan. I was looking for a specific combination of things: a good research fit, a program that did not require me to spend five or more years away from industry, a realistic way to keep working part time, and a place where I would actually want to live.

Japan happened to fit my constraints unusually well. There was a lab in Tokyo doing research that matched my interests, the Japanese PhD system pushes students to graduate in three years, and the visa situation was more flexible than what was available elsewhere.

On top of that, I already knew that I liked dense, walkable megacities from my time in Seoul and New York. Tokyo was the kind of place I could imagine living for several years.

There were also less concrete reasons. Time in Japan is a good conversation starter and door opener. It sets you apart from other applicants and shows that you can take on challenges and do things differently. I do not think this should be your main consideration when choosing a PhD program, but if you are weighing all the options, it is not irrelevant.

Why Japan can be a good place to do a PhD

Overall, Japan has a strong academic culture, especially in STEM due to the University of Tokyo, Kyoto University, Tohoku University, and my alma mater Institute of Science Tokyo. While undoubtedly the most prestigious schools are disproportionately located in North America and Europe, Japan also has some heavy hitters on the world stage. But this can be said about many countries; in Asia specifically, China, Korea, and Singapore each have several top-tier schools.

A flexible student visa

Japan had other qualities that drew me to it. A practical one is that Japan has a relatively lax policy for working while on a student visa.

Japan’s student visa can come with work authorization from day one, provided you file the correct paperwork either in person or online

This work authorization is subject to only mild constraints, like a cap of 28 hours per week during the semester and 8 hours per day during long school breaks, as well as the exclusion of some types of work.

The ability to work part time is an incredible benefit of doing a PhD in Japan that is rare elsewhere. In America, international students can do internships, but not sustained, year-round work. American students can work freely, but it is rare to do more than summer internships and occasional consulting.

Other countries also seem to have strict requirements. For example, in Korea, graduate students on the D-2 visa are only allowed to work part time in limited roles and only after receiving permission from the school and government. Even summer internships in Korea are much more difficult to arrange. While some students I know in Korea would travel back home or to another country for a summer internship, this came at a large cost, to say nothing of the difficulty of traveling long distances. Since computer science and machine learning blend academia and industry so well, and the incentives for joining the industry after graduating are so high, not being able to do industry internships is a major setback for students.

A PhD with a deadline

Another practical point is that, unlike the US and Europe, which often require five or more years to complete a PhD even with a related master’s, the Japanese PhD system pushes you to graduate in three years. The opportunity cost of doing a PhD is high, and the advisor-student relationship can easily become unbalanced. Students often pay the opportunity cost of being out of the workforce for many years, while labs benefit from their research output and are incentivized to keep the student as long as possible, potentially by delaying their graduation. In that context, having a structural time limit is helpful.

However, the downside to this is that three years can feel too short to be able to reach the edge of your field, produce new work, and train as a researcher. In fast-moving fields like machine learning, one can become productive very quickly, but having a shorter PhD time limit pushes students towards less risky and innovative research to mitigate the risk of taking a bet on a research topic, having it fail, and then not having enough time to recover and produce good work before you need to graduate.

A quickly-growing tech scene

A final good reason to do a PhD in Japan, at least in machine learning, is that the tech scene has changed dramatically and offers opportunities to connect with top companies.

When I first looked into moving to Japan, I would not have considered staying in Japan to work at any company other than FAANG. By the time I finished, OpenAI, Anthropic, Cohere, Hugging Face, and Sakana had opened offices in Japan, which made it feel much more viable as a post-PhD destination.

Being already in Japan gives you a leg up in connecting with these companies, and insulates you from a lot of the international competition that you would face in Europe or America.

Why Japan can be a bad place to do a PhD

There are many downsides to doing a PhD: the opportunity cost, increasingly scarce faculty jobs, the publish-or-perish mentality, etc. Likewise, there are downsides to living in Japan: being far from home and loved ones, relatively low pay, social isolation, etc.

On top of this, the combination of doing a PhD in Japan comes with an additional set of challenges that prospective students should be well aware of.

You went to what university?

The first is name recognition. Doing a PhD abroad doesn’t always have to come with a name recognition penalty, but if you travel outside of your immediate geographic area (e.g. outside of America as an American, or Europe as a European), you quickly run into name recognition issues for all but the top, globally-recognized schools.

Many American universities have good name recognition worldwide, but I think you would be harder pressed to find people that could accurately list the best schools in Korea or Japan if they had not lived there—compare this to the generally high name recognition of Cambridge, Oxford, EPFL, or ETH Zurich to non-Europeans.

Imagine you are applying for jobs later and the person doing the initial resume screen has never heard of Tohoku University—which I mention specifically as I think people would assume that “University of Tokyo” was a good school, even if they did not know about it directly, but Tohoku University is less well known outside of Japan, despite it being a phenomenal school. You will likely be passed over in favor of candidates with more familiar credentials, even if they come from lower-caliber schools.

To put this another way: assuming you’ll be a good graduate student either way, my opinion is that the expected outcome of doing a PhD at a good-but-not-top-tier university in your home country is better than doing it at a good-but-not-top-tier university abroad (with the exception of the US, whose mid-tier schools still carry relatively high name recognition worldwide), if your goal is to move back to your home country after graduation.

Of course, name recognition isn’t everything, and doesn’t actually capture how good your university or lab is or how strong of a researcher you are, but when dealing with high variance situations like hiring, having good name recognition eliminates a huge potential downside.

Unnecessary courses

Another downside of the Japanese PhD system is your academic course load. At Tokyo Tech, PhD students were simply unable to take graduate-level technical courses. This is in stark contrast to most other PhD programs I know of, where extremely niche technical courses are available to PhD students, who are able to get consistent, focused face-time with experts in their field that are not from their lab.

On the one hand, this is a good thing. You have three years to finish your PhD and you should spend that time doing research, not taking courses. But in order to graduate, you still have to complete dozens of credits. At my school, these were divided into seminars (effectively free credits just for doing research), humanities, and career development courses.

For the humanities, I was able to take several Japanese courses, which were fun, though not very effective due to the class sizes. However, the career development courses were not useful. Though I am a native English speaker, I was required to take English presentation courses, which were not effective in improving students’ English abilities at all.

I was also required to enroll in several “future career plans” courses that were basically a series of hour-long speeches by alumni. There was very little interaction in these, so even though many of the speakers had interesting backgrounds, it was hard to get any good discussion going. So while this may have served the purpose of introducing potential career paths, it did not allow for any detailed back-and-forth nor did it serve the (presumed) secondary purpose of being a networking event for students.

You published where?

A final issue I saw more of in Japan was the emphasis on domestic journals and conferences for publishing. There is a well-established set of domestic venues for many fields, and students are often encouraged to submit to them. These publications usually appear in Japanese and may have much lighter peer review than major international venues, but they can still count toward graduation requirements, so many students take advantage of them. However, these conferences and journals are not known outside of Japan.

Just being a domestic rather than international venue means that most employers will discount any research published there, since there is no way to easily verify the quality or prestige of the venue. In some cases they can be useful for local networking or practice, but I would be cautious about relying on domestic journals as your main research output if you plan to leave Japan after your PhD.

Where do you want to be after the PhD?

Looking back, I believe the key question to ask yourself before you commit is: where do you want to be after your PhD? The core question about doing one at all, “Is it worth the opportunity cost?” is the same regardless of where you study. Likewise, most questions about whether Japan suits you apply to anyone moving there, not just students. So to me, the decision point comes down to your plans for after your PhD.

If you think you want to stay in Japan after your PhD, the questions you need to ask aren’t unique to being a PhD student, but are the same for nearly everyone who comes to Japan without a set end date:

• Are you okay living in a country that’s difficult to fully assimilate into, no matter how long you stay?
• Are you okay being far from friends and family?
• Are you okay with the pay, and especially with how hard it is to reverse course if you eventually decide you want to move home after all?

If you can’t answer most of these with a confident “Yes,” you should think twice about your future plans.

But, if your plan is to do a PhD in Japan and then leave, your list of questions becomes much shorter. The primary debate is whether you’re willing to risk the loss in name recognition when you return home and compete against peers with household-name credentials. I personally think this is a risk that people do not take seriously enough, especially if you choose a school outside of the top few schools in Japan.

If you are a strong student and do good research, the hope is that your merits should speak for themselves and potential employers will recognize them. But I do not think this is something you should rely on happening, especially in an increasingly competitive and uncertain job market. This potential downside of choosing Japan can be so great that it outweighs all of the good things about Japan, and so it should be considered carefully.

In short, you are stuck with a hard choice: move back home after graduation, or commit fully to Japan. The latter option is attractive in Japan, as evidenced by the number of “lifers” that you will encounter, but it is still risky. You are tying yourself to a culture that is difficult to integrate into, and the longer you stay, the more difficult it becomes to reverse course.

Finding the right lab

The critical first step to getting a PhD, in Japan and elsewhere, is finding a lab and a professor to work with. There should be a research fit (it can be loose, but there should be some meaningful connection) and a personality fit (do you have the same work style, does your professor micromanage when you want hands off, etc.). Each of these greatly narrows down the prospective field of professors.

Add to this that you are restricting yourself to Japan—which admittedly has many good universities to choose from, but it is still just one country—and that the research and personality fit has to go both ways. Suddenly, you have very few options.

Then there’s the fact that Japan is a desirable country to live in for people all over the world, so professors are inundated with prospective student emails. You have to stand out from a crowd of other, presumably highly-qualified, applicants, and they must make sure you are a serious student and not just looking to join their lab in order to be in Japan.

How I did it

Over time, I had identified a few labs that might be a good fit for my research and other criteria. I only considered Tokyo, so this really narrowed things down. I read up on the current students’ research, what the professor was working on, what the trend of the lab was, etc. I was looking for labs with many international students, a hands-off professor, and research similar to my intended area in Japanese, Chinese, and Korean language modeling.

I happened to meet a graduate student from my eventual lab at a conference and had an in-person chat and some follow-up emails with him. I also had a friend that was a graduate student in the same department, though in a different lab. He was able to put me in contact with another graduate student from the lab. I managed to briefly visit the lab on a business trip to Japan before COVID-19, where I arranged to meet with the professor for an informal meeting.

These interactions allowed me to scope out the lab and the professor, establish a rapport, demonstrate that I was serious, and discuss the possibility of being a graduate student there. There was interest on both sides, but I was still early in my career, and I was not ready to leave my job yet.

Later, as I passed the two-year mark at my work, I began to arrange my proposal to switch to part-time work and a full-time PhD. My managers were largely on board and this ended up going pretty smoothly.

I reached back out to the professor to tell him I was applying. We exchanged some more phone calls and held a series of research interviews, which more or less sealed the deal.

Funding the degree: MEXT and the three-year clock

I knew applying for the MEXT Scholarship would be a good next step, both for the prestige and also to help offset some of the costs of moving to and living in Japan.

Few options for financial assistance

The MEXT Scholarship is one of the main funding paths available to international graduate students, and I would not recommend doing your PhD in Japan if you do not have it or something similar. Self-funded PhD students are common in Japan, so universities do not always have robust scholarships available, and you will likely not be paid by your lab—at least, not an amount that would be comfortable to live on.

Working part time can alleviate financial issues, but it is better to have your funding set up unconditionally, rather than adding an additional stressor onto an already stressful PhD. All of this is significantly different from doing a PhD in the US or Europe, so it may come as a surprise to those considering Japan for graduate school.

For some context, the MEXT scholarship generally grants tuition support, airfare to Japan, and, for doctoral students, a monthly stipend of about 145,000 yen (approximately 900 US dollars at the time of writing). This is enough to live on, but not enough to completely remove money as a source of stress, especially if you live alone in Tokyo.

The two flavors of MEXT Scholarship

Embassy Track

Japanese consulates are allocated a few scholarship spots and hold interviews for candidates within their districts. This track is extremely competitive, as consulates are given very few openings. It also technically comes with you foregoing your ability to choose your eventual university and laboratory. In the application, you are allowed to rank several universities and laboratories as your top choices, but you aren’t guaranteed to get placed at any of them.

I took the interview in Atlanta and bombed miserably. In retrospect, I had a weak story for why I wanted to go to Japan. What I said was true: I was interested in Japanese culture, I had found a lab that did what I wanted to do, I had a strong research background and experience living abroad, etc., but it just wasn’t very compelling overall. I think it looked like I was just someone who wanted to take a break from work and live in Japan for a few years, rather than someone who was striving for it.

University Recommendation Track

After the consulate interview results came out, my future advisor suggested I apply for the other MEXT track: the University Recommendation Track. To be honest, I did not know that this existed at the time, and if I had, I would have just done that in the first place.

This track involves first applying to the university of your choice directly, and once you are accepted, asking them to refer you to MEXT. You typically have already come to an agreement with a specific professor to join their lab, and they write in support of you.

I am not sure how many scholarship spots each university is allocated in this program, but my guess is that it is simultaneously harder and easier than the Embassy Track. It is harder in that the bar to apply is much higher, since you must have a professor argue on your behalf; however it is easier in that if you have a professor’s strong support, you are very likely to receive it. This is unlike the Embassy Track where you have significantly less agency over the situation, and it is a good choice if you have a strong research fit with or connection to a professor.

The three-year limit

The MEXT Scholarship only guarantees you funding for a three-year period, with limited opportunities for extension. This strongly couples with the norm of completing a PhD in Japan in three years, and imposes a more tangible deadline on your graduation.

You can usually extend your time as a graduate student by several years just by filling out some forms and talking with your advisor. But, if the money runs out, you are left in a bad situation.

In my particular case, I was a student for six months beyond my MEXT allocation due to bureaucratic issues with my PhD thesis submission, and because I began my PhD by taking a leave of absence for a semester since I was not yet able to enter the country due to COVID-19 (but this time still counted towards my MEXT scholarship). I wasn’t taking any classes or even really required to be at the university at all, but since my MEXT had run out, so too had my tuition waiver. Not only was I no longer getting my MEXT stipend, but I was also on the hook for my tuition fees which, even though they are an order of magnitude less than US tuition, were still non-negligible. I had the money to cover this, but someone who was relying totally on the MEXT scholarship for their finances would have been in a difficult spot.

Working part time during a PhD

Overall, I would recommend trying to work part time while doing a PhD if you are in Japan, but only if you already have a strong research background and some industry experience. You need some time to ramp up your research/engineering skills and saturate your research pipeline so you always have something to work on. If you aren’t already comfortable in these areas, you will end up “half-assing two things instead of whole-assing one thing,” which will be extremely bad for the outcome of your PhD and could burn a bridge with your employer.

My specific situation in Japan was somewhat special. I transferred from being a full-time employee at Google in the US to being a part-time PhD Researcher in the Tokyo Google office working two days per week. I’ve received a lot of queries about how it worked and how one can get a similar role, but overall, this is not a very repeatable situation. It required the Japanese government, my lab, and Google to all be okay with the arrangement, and it helped that my Google team happened to have a subteam in Tokyo that I could immediately transfer to.

Having the government on board by default makes the process much easier, but finding a position is still non-trivial, though my feeling is that (especially smaller) companies are starved for hyper-niche technical talent, so it is a win-win for them to hire someone part time to get access to your skills at a reduced cost.

Often, then, the most difficult aspect is getting your advisor on board. While they cannot actually prevent you from working part time if you want, other than by following through on a threat to kick you out of the lab, your advisor has an immense amount of influence over your immediate future, so you risk a lot by going against them. My advisor, thankfully, was supportive of it, though it would be different if the student started doing poorly in their research. Not all advisors are this way.

Still, I know of many people in Japan who worked part time at interesting startups throughout their entire PhD. This is a very viable path that can greatly boost your quality of life, assuming you can juggle work and research, as well as give you a leg up after you graduate.

Social life at the university

At the lab

Despite my lab being fairly international, there was a clear divide between the Japanese and international students. Everyone got along and there were no major issues, but the separation was tangible. This mainly manifested as a divide over who collaborated on projects together, with international students primarily working with international students and Japanese students with Japanese students.

This makes sense in some ways, as the language barrier can be hard to overcome, especially in hyper-technical contexts, but it is not ideal. If there are not many international students in your laboratory and your Japanese skill is not sufficient for technical conversations (like mine), you may find yourself with few immediate collaborators.

As a PhD student, it is important to learn how to network and find people with whom you can conduct good research, but as a beginning PhD student in particular, it is helpful to have a supportive set of labmates who you can work with and learn from.

I ended up becoming pretty good friends with some of my labmates and we would hang out outside of school. This helped a lot to ease the isolation that many people feel in Japan, since labmates are basically primed to be friendly in a way that coworkers are not. Our lab also hosted semesterly dinners, game nights, and other small events to help break the ice.

No sense of community

One stark difference that I noticed in Japan, compared to the US and Korea, was the divide between graduate and undergraduate students—or, perhaps, the divide between a graduate student at a lab, and the rest of the student body.

In the US, I felt there was very little barrier between me and the graduate students in my department. I saw them in passing and knew some by name. They taught my classes, hosted and attended seminars, etc. In Korea, there was a greater divide, but I still interacted with undergraduates often through seminars, courses, general events, and students joining our lab.

However, in Japan, I felt that graduate students and undergraduates were essentially two different species that did not interact. I struggle to think of a single time I talked with an undergraduate student who was not part of my lab. There were no opportunities to be a teaching assistant (which some people dread, but I was looking forward to) and no random hallways in which to bump into people.

Even interacting with other graduate students outside of my immediate sphere was difficult, as my lab was behind two key-access doors on top of a random building on campus. This was quite different from in the US and Korea, where the lab space, classroom space, and general departmental student body space were all well integrated.

I found this a bit of a bummer. Of course, by the time I was a graduate student in Japan, I was nearing 30, and so probably would not have made many friends with the much younger general student population. Still, it made my time at Tokyo Tech feel more like rote work and less like being in a thriving academic community.

Overall, while my lab experience was very nice, the university itself offered little else in terms of socializing. You cannot rely on the crutch of being in school to build a social network the way you might as an undergraduate, so you have to be proactive about cultivating relationships outside of the university.

Living in Tokyo

My daily life in Tokyo was quite different from my experience in big American cities, but similar to Seoul.

Making friends

Being an international student in these countries means you are, de facto, an outsider. A common sentiment among foreigners in Japan is that even if you speak perfect Japanese, have a Japanese partner, work in a Japanese company, adhere perfectly to Japanese customs, etc., you will never fully fit in. I don’t speak perfect Japanese, nor do I have a Japanese partner or employment in a Japanese company or a perfect grasp of Japanese culture, and I very clearly felt the gap while living there.

The Japanese friends I made were all either from my lab, or native Japanese people that had lived abroad. Overall, my friend group was dominated by expats. Many people may interpret this as defeating the purpose of living in Tokyo and fully immersing in Japanese culture, but it is nearly a fact of life. As I explained above, you should not expect to make a huge group of Japanese student friends at your university as a graduate student. Fortunately, the expat scene in Tokyo is large and diverse, at least relative to Seoul in my experience.

I tried a few different ways to meet new friends, like attending tech meetups or even Bumble BFF, but in the end I met my closest friends during chance encounters at coffee shops or on the street. Even in my home country, encounters like this are difficult to come by. In Japan, especially if you are in Tokyo, the situation is even worse, since the likelihood of meeting a foreigner that is not a short-term visitor is very low and making Japanese friends is difficult.

All this is to say, if you don’t already have a support group in Japan, you must work very hard to deliberately build your friend group.

Housing

My choice of housing had a few pros and cons. I prefer to live in very dense areas of a city and am happy to pay a premium for a short commute. I chose to live just outside of Shibuya for proximity to work and my lab. A nice benefit of this was that I commuted against traffic, so I largely avoided crowded trains.

I found a pre-furnished apartment and stayed in that place the entire time. The convenience was helpful, especially because I could lease it while still abroad to avoid scrambling for housing when I first arrived, avoid the biases of renting in Japan as a foreigner, and not pay key money. But because of this, I paid a premium.

This is an important factor for those who want to do a PhD in Japan. The MEXT scholarship can provide enough money to get by, but you will have to sacrifice a lot of quality of life, and not having the scholarship will mean you are paying everything out of pocket while not receiving any income as a student.

Supplementing your income with a side job in tech therefore becomes a very attractive situation, as you can simultaneously massively improve your quality of life, and gain experience and insight that you cannot get as a student.

Daily routine

I settled into a routine pretty quickly in Japan. During the workweek, I went to my lab or workplace depending on the day, and usually stayed until fairly late. This wasn’t enforced by my lab: I just genuinely loved doing my research and stayed late to work on it. I would usually come back to Shibuya for a late dinner or cook something small at home.

On the weekends, in the beginning I would spend one day at the lab—which was also not enforced, but rather, gently discouraged by my lab. The other day I’d explore some part of the city or try to meet up with friends. I eventually got comfortable working remotely and stopped going in on weekends.

Around my second year in Japan, when the country was open to visitors and my partner was able to visit more, I began to travel around the country. I enjoyed many trips by getting a good work session in while on the train to whatever city, spending a few days there exploring (and still working, of course), and then having another good work session on the train back to Tokyo. The convenience of Japan’s transportation network, combined with the flexibility of being a PhD student, made this kind of working holiday easy.

In a reasonable lab, especially in computer science, there’s no real reason to be physically present if all you need is internet access. You should have considerable leeway to work when and where you see fit.

After graduating

Near the end of my PhD, I began exploring my post-graduation options. I was very comfortable in Japan and enjoyed my life there, but the fact is that being an expat is difficult. Both my family and my partner’s family live in the US, and this puts a lot of pressure on us to be closer to home.

Additionally, while my partner enjoyed Japan, she was not as comfortable there as I was due to the language barrier and the difficulties of finding work in her field. Software engineering provides a lot of leeway in terms of location, remote work, etc., all with relatively high pay, but my partner’s field, healthcare, is much harder to break into for non-natives.

Coupled with the uncertainty of where we would eventually want to retire and the pay differential, the benefits of moving back to the US outweighed my desire to stay in Japan. Ultimately, I did receive comparable offers between Japan and the US relative to cost of living, but the access to a 401k and the higher pay tipped the scales.

Epilogue

Having now moved back to America and (re)settled into my routine, I have mixed feelings. There are good things about moving back. I am closer to family, I enjoy my work and am able to save money, my partner is happier, etc. Still, I cannot help but miss Japan terribly.

In essentially every other category, my quality of life has degraded significantly since returning home. Whereas I loved the energy, walkability, safety, cleanliness, and variety of Tokyo, I am now in the suburbs of Silicon Valley, which simply cannot compare.

San Francisco and the Bay Area have incredible things to offer: nature; droves of interesting, smart, and ambitious people; being the epicenter of technology, etc. However it is lacking a certain magic that makes Japan so interesting.

I find myself thinking about moving back, and what opportunities I could find. Overall, I think I made the objectively correct choice, but with decisions like this the subjective aspects make everything more difficult.

Re-reading this article, I feel I emphasized the negative aspects of Japan more than the good. Part of that is intentional, as people often see Japan with rose-colored glasses, and I think it is important to be honest about the downsides. But a lot of what’s good about Japan is intangible, and harder to put into an essay like this.

So I will still conclude by encouraging people to take the risk and try out Japan if they are considering it. I can almost guarantee it will be a life-changing experience.