From Blog to Job Board: How Twenty Years Inside Japan's Software Industry Taught Paul McMahon What the Market Gets Wrong

Interviewed by Fuminori Gunji, Head of GTM, ZooKeep

Paul McMahon is the founder of TokyoDev, a job board and community that helps international software engineers find work in Japan. A Canadian computer science graduate, Paul arrived in Japan in 2006 on a working holiday visa with minimal professional experience and no plan beyond trying something different. Over the next two decades, he co-founded a mobile development consultancy, built and scaled Doorkeeper, one of Japan's most widely used event management platforms, and eventually turned a personal blog and informal mailing list into one of the most trusted resources for international developers navigating Japan's tech job market. In this ZooKeep interview, Paul speaks with ZooKeep Head of GTM Fuminori Gunji about the historical forces that shaped Japan's software talent shortage, what it actually takes for a Japanese company to successfully integrate international engineers, and why he thinks the market may be closer to turning a corner than most people realize.

Human Origins

Vancouver, a math-turned-entrepreneur father, and the quiet seed of a builder's mindset

Paul, let's start at the beginning. Where are you from and what was your family like growing up?

I was born in Vancouver, Canada. My father was an entrepreneur. He commercialized his math master's thesis from university, where he was doing research into using computers for manufacturing silicon wafers. This was in the early eighties. There wasn't much being done in that area, and it naturally turned into a business. He exited when I was a teenager. My mother was a professor of art history and taught until I was around five, then became a stay-at-home mother with me and my two sisters.

That's quite a combination: an entrepreneur and an art historian. Did either of them push you toward a particular path?

Not consciously. I never felt any pressure. But I do remember one time when I was a child, I was telling my father "someone should invent this thing," and he said, "you could invent it." That idea, that maybe I can do something new too, just planted a seed. I didn't recognize it at the time. But looking back, I can see the entrepreneurial side coming from him, and from my mother I get the more human side of things, caring about people, caring about social justice. TokyoDev ended up being a kind of amalgam of both: doing business, but doing it with a purpose.

How Japan Happened

A community college detour, a co-op program that didn't pan out, and a Japanese roommate who changed everything

You arrived in Japan in 2006 on a working holiday visa. A Canadian computer science graduate who hadn't worked professionally yet. How did Japan of all places end up being the destination?

It actually connects back to university. I originally went to a community college in Vancouver without knowing what I wanted to do. I took random courses, English, political science, math. Then I chose a computer science course almost by accident, and it really clicked. I got motivated, transferred to another university, and when I got there a counselor mentioned there was a co-op Japan program. Around the same time, I was living in a shared house where one of the other students had done working holiday in Japan, and one of my roommates was Japanese and doing a working holiday in Canada. The timing of the official co-op program ended up not working out for me, but I realized, why not just try it on my own?

Did you consider other countries?

Not really. Japan was already in my mind from the co-op program. I had met some Japanese people, it seemed interesting, different from Canada in terms of how people think and the culture, but also approachable, on a similar socioeconomic level. It didn't seem like a scary place to go. I took first-year Japanese for extra credit at the end of university, then came here.

Early Career in Japan

A startup with a broken business model, a consultancy built on SEO and Japanese feature phones, and the moment Doorkeeper was born

Walk us through those early years.

Right before coming to Japan I was looking online for jobs and there weren't many opportunities for a junior developer without real professional experience. But I found one job ad that was clearly written by a software developer, not a recruiting agency. It actually spoke to me as a developer. I emailed them before I even arrived, set up an interview, and managed to get the job. I joined this Japanese startup where the development team was almost entirely non-Japanese, around ten software developers in total. I learned a lot from them. But the company's business model didn't really make sense. After two and a half years, I left with a couple of colleagues.

We weren't sure what came next, so we started a software development consultancy called Moblin. One colleague wanted to do freelancing, my other partner and I thought, why not try this? We didn't have anything specific in mind. We just knew the current situation didn't work for us.

How did you get clients? You were all foreigners.

Our clients came from a few places: our professional network, former colleagues, attending industry events and meeting people there, and SEO. We'd chosen to specialize in Japanese feature phone mobile development. This was pre-iPhone. Japanese flip phones had built up their own sophisticated way of doing things, and from our previous startup we had knowledge of that. We put out some material in English about how to develop for it and managed to land Estée Lauder as a client. We were all software engineers, so doing content marketing rather than direct sales made more sense for us anyway. And I guess that inbound approach has been my pattern ever since.

While running Moblin, we had a side project: Doorkeeper. It started as a check-in solution for Mobile Monday Tokyo, a networking event we were involved in. Then we decided to generalize it so other communities could use it. We announced it on Twitter. Nobody used it for the first month. Maybe one person tried it. But we kept at it, stayed involved in other events, in the startup community and the Ruby programming language community, and after about a year our connections started trying it. By around 2013 it was the most-used event platform in the Japanese tech community.

When did it become a real business?

That was the problem. Our business model was transaction fees: ninety-nine yen plus two point five percent of paid ticket sales. We were processing over a hundred million yen a year in ticket sales and only making a small fraction of that. Many of our events were free anyway. So the business model didn't work.

After several years, I was ready to give up on it. My co-founder had already left Japan and we were going to shut it down. But before doing that I thought, why not try a subscription model? I did the work to integrate subscriptions, and wrote an article explaining the thinking behind it. A lot of our users left. But enough stayed that it was finally generating enough revenue for me to pay myself a decent salary. I ran it that way for a couple more years, but revenue plateaued. Meanwhile, my personal blog, TokyoDev, has been growing on its own.

How TokyoDev Became a Business

A blog, a mailing list, and a discovery that people were getting hired before there was even a product

TokyoDev started as your personal blog. When did it become something more?

I'd been writing about being a software developer in Japan since around 2010. Just for myself. People kept coming to me asking how to find a job here, so I wrote articles about it. I started a mailing list where I'd send out job opportunities I came across, not systematically, just things I noticed. Then I started hearing from people that they were actually getting hired through it. And I started hearing how much recruiters were charging in this space. Something clicked. Maybe there's a business here.

By around 2015 I started charging companies a success fee if they hired someone through us. It wasn't much at first. But by 2019 or 2020 it was generating more revenue than Doorkeeper. I exited Doorkeeper, sold it, and since 2021 TokyoDev has been the only thing I've focused on.

What is TokyoDev 

A job board, a media business, and a community that serves its tribe first

TokyoDev is a job board and community focused on helping international software engineers find jobs in Japan. This could be people already here or those looking to relocate. We focus on companies that offer an English-friendly environment, where they're looking to hire good software developers without necessarily requiring strong Japanese skills.
Founded by Paul McMahon as a personal blog in 2010, it evolved into a job board with a success-fee model around 2015 and has since become the primary resource for international developers navigating Japan's tech job market. TokyoDev also conducts an annual survey of international developers in Japan covering compensation, working conditions, technology stack, and AI usage, now in its fifth year with nearly 1,000 respondents.

You describe it as a job board, but looking through the blog it feels like much more than that.

TokyoDev wasn't a business in the beginning. It was my personal blog, driven by wanting to help my peers. That's always been the primary focus. The job board is our monetization strategy. But it's just part of what we do.

With the articles, I want to have interesting content on a wide range of topics. There are areas where there isn't enough good information and I feel like there should be. We've been opening it up to guest writers too. I just put out an open call for contributors and was very transparent about our process and what we pay. In two weeks, forty people reached out. Not all of those pitches worked, but it gave us a stream of perspectives I never would have found myself, and it keeps me connected to what people are actually experiencing. I've been in Japan for twenty years. It's been a long time since I was the one looking for a job. Staying close to the community's real challenges matters.

The SI History Behind the Talent Gap

Why Japan's system integrator era didn't just shape how software got built, it shaped who got good at building it

Let me ask you something I think a lot of people wonder about but don't fully understand. What is actually behind Japan's software talent gap? Not the labor shortage headlines, but the deeper structural cause.

When I arrived in 2006, almost all software development in Japan was done by SIer companies, system integrators. Not companies building software for themselves, but companies hired by others to build custom software. And these SIers didn't really have an incentive to hire the best people, because they billed by the hour. Their revenue came from how good they were at selling and managing the relationship, not from how good the product was. Whether what they delivered was effective for the customer wasn't really their problem.

For the English-speaking audience, let me describe how that software development process looked like:

Fuminori: The process was essentially a spreadsheet exercise, I was on the customer side during my time at SoftBank working with a typical Japanese SIer. You are given a specification spreadsheet with columns like Requirement ID, Category / Sub-category, Requirement Name, Requirement Description / Details, Priority / Importance, Remarks / Preconditions, Target Business / Use Case, Actor / User Permissions, Input Information / Trigger, Output Information,  to describe your business requirements to the solutions engineer or project manager (not a product manager) of the SIer company. 

The project manager will then together with their engineering team interpret and translate those into business requirements into system requirements and send them back to the typically non-technical customer for sign-off (which is very hard to even understand or imagine how the UIUX might look like) but after a couple of 2 or 3 hour meetings mostly relying on text based information only, the customer has signed off at some given deadline. 

Then the SIer would say, "We'll build exactly what's written here." And they did. No guarantee it would be effective or useful. No common sense applied, just literal interpretation of the spreadsheet. 

If the customer journey turned out to have too many clicks, or is cumbersome to understand what to do next for the user, the SIer would simply say "that's not in the spec sheet" and fix it a pre-defined number of times after which they would start charging you extra for asking for more changes.

In a perverse way, the worse the UI design, the more money they could make, because every additional fix was billable. 

What do you think was the knock-on effect of this market environment on the engineering talent landscape in Japan?

When software development is treated as something anyone can do, companies hiring new graduates and giving them a three-month course regardless of aptitude, the typical level of software developers stays low. Which then means companies don't value it. Which means why would anyone who's good want to do it? Low salary, long hours, being told what to do with no room to think. It stifled the industry.

It also goes against how most software is developed internationally, where it takes a more agile, collaborative way where engineers are incentivized to think end to end and use common sense, push back on senseless requirements, question the customer etc. For that to work, the engineer needs a certain status. They need to be able to talk directly to the customer or whoever the users of the software are going to be. The SI model actively prevented that.

When did you start to see a change?

Around 2011 and 2012, Japanese startups started popping up who were building their own software products. When you're building your own product, you care about the long term. You need engineers who are good and who care about what they're doing. I saw this very clearly inside Japan's Ruby developer community. When I first started, those developers were passionate about Ruby but stuck in SIer company jobs where they weren't even getting to use that programming language. Within a few years, all those same people were becoming CTOs of new startups, finally surrounded by others who shared their passion for the craft. As software developers became more important within companies, those companies started looking for best practices globally. And when they couldn't find senior talent domestically, they started looking outside of Japan.

The International Engineering Talent Market: What the Numbers Say

Under one percent of developer jobs require no Japanese language skills in Japan, and the companies that do offer them are almost by definition the best ones

TokyoDev runs an annual survey of international developers in Japan covering compensation and working conditions. What were the most interesting findings?

The survey started because people were talking about how the typical market conditions for a software developer in Japan are not great. But that wasn't the experience I was seeing among my peers. I wanted to highlight what it can actually be like as an international developer here, and so we started conducting this survey annually.

What percentage of software developer jobs in Japan don't require Japanese language skills?

Probably under one percent. There are more companies willing to hire someone with business-level Japanese, where they can do their day-to-day work even if it isn't perfect. But truly English-only opportunities are rare.

And what do the companies that do offer those roles have in common?

They've evolved as software development has changed in Japan. The early adopters were startups building their own products. More recently I also see Japanese founders who want to build a global company from day one. They recognize the Japanese market can be a trap, it's big but it's not a high-growth market. So some of them are building international teams from the start, partly because they believe it will help them become genuinely global.

When Integration Fails and When It Works

The two-cultures trap, the bilingual champion thesis, and what developers actually complain about

You must see a lot of cases where a company hires an international developer and it doesn't work out. What does failure look like from the outside?

I think for such integrations to succeed, you need someone at or near the executive level, or at least someone with a strong relationship there. It's not just hiring international people and expecting them to adapt to your organization's way of doing things. You have to change as a company too. And there needs to be someone who is empowered to make those changes.

The cases I've seen where companies successfully move from a Japanese-only team to one that incorporates international members almost always have this kind of leader. Usually a bilingual Japanese-English speaker with experience in both Japanese and international companies, who can bridge the gap and actually take action.

What frustrations do international developers most commonly report?

Compensation is the biggest complaint, especially with how the yen has moved (as of July 2026, 1 USD = approximately JPY 162, near its weakest level against the US dollar since 1986). Beyond that, I see companies end up building two cultures instead of one. They'll build a strong international software engineering team with good internal relationships, and a strong Japanese business team with good internal relationships, and then there's supposed to be some way they work together. The more common approach is to have bilingual people act as bridges. But that just introduces an extra layer. The bridge people get overwhelmed. People on both sides don't feel like they can communicate directly.

The best companies find ways to get the business side and the engineering side, the Japanese and the international, to communicate directly. AI and translation tools are making that easier in text-based communication. But human relationships matter too. Some of our clients do offsites or company-sponsored lunches specifically structured so people have to sit with someone from a different team. Those are the organizations trying to build something coherent.

The Gender Pay Gap

Why it doesn't exist early in a developer's career in Japan, and why it widens sharply after that

You've been transparent about finding a gender pay gap in your own survey data and then taking action by sponsoring organizations like Rails Girls Japan and Women in Tech Japan. What's your read on why the gap exists, and what's missing from the usual public debate?

I'm not an expert in this topic. What I felt I could do when I saw it in our data was support the communities already working on it. But one aspect I've seen personally that doesn't get talked about enough is how cultural norms indirectly push women out, or require them to make sacrifices men don't. Nomikai culture is a good example. If a company has a drinking event after work, it's much more likely the father will attend than the mother, because someone has to take care of the kids, and that defaults to the mother. None of that is explicitly discriminatory. But it compounds over time.

Overtime has the same effect. If your company allows or expects overtime, you are penalizing mothers more than fathers just because of how society is already structured. You're not actively discriminating, but the outcome is still there.

If we really want to change things, men need to make sacrifices too, share the load of raising kids so that women can maintain their careers on equal footing.

The Future of the Market

AI layoffs in Silicon Valley, no slowdown in Tokyo, and why translation tools don't solve culture

Japan has a well-documented labor shortage. International engineering talent demand is growing. But the actual numbers of international developers being hired and retained are more complicated than the demand signal suggests. What do you think will happen in the next five years?

The main obstacle is Japanese companies themselves and their ability to adapt. It's genuinely hard to go from an environment where everyone shares the same high-context language and culture to one that doesn't assume those things. You need internationally minded leaders within these companies who have experienced both environments, and you can't manufacture those people overnight.

You've said TokyoDev hasn't seen a hiring slowdown, even as AI-driven layoffs have hit developers hard in the US. Do you expect that to change in Japan?

We haven't seen it yet. In the US there's been significant pullback, including for senior people at large tech companies who are strong and now struggling to find something else. Will that come to Japan? We'll see. But even now, people from that background who've been wanting to move to Japan have struggled to find roles here, simply because there are so few English-speaking positions available.

The biggest opportunity AI creates in the near term is lowering the language barrier to direct communication. That helps with the two-cultures problem I described earlier, at least in text. But it doesn't solve culture. And culture is what actually determines whether integration succeeds.

Is there a message you'd put on a billboard for Japanese software engineers right now?

We're trying to encourage more Japanese software engineers to become globally minded, to take on the challenge of working in a more international environment. There's an option they don't always see: it doesn't have to be "move to Silicon Valley or stay in a Japanese company." There are companies in Japan that have already built international software teams. Those are a great starting point, a softer on-ramp to a more international career, while still being in Japan.

Why ZooKeep

On education-first marketing and what Paul noticed about ZooKeep before any product conversation happened

You mentioned ZooKeep in a LinkedIn post about ATS frustrations in Japan. What drew your attention to us specifically?

With ZooKeep it's been less about the product and more about the people. You seem more focused on education and helping people rather than directly selling to them. That's why I've been happy to do events together and to introduce our customers. I knew it wasn't just going to turn into a pitch session. Jordan's presentations don't even really mention ZooKeep directly. That way of selling matches how I've always approached things, and it's refreshing in a space where so many people sell too hard, too transactionally, too focused on getting something back immediately. That isn't what I've seen with ZooKeep.

Carve Outs

Recommendations from Paul McMahon

Is there a book or resource you'd recommend to someone starting out, either as an entrepreneur or as an engineer trying to navigate this market?

Earlier in my entrepreneurial journey, The Lean Startup by Eric Ries was really helpful. There's a case study on Groupon in there, where the whole point is that Groupon started as just a blog with a sign-up form. The idea of not actually building a product until you really need to, was how I approached TokyoDev: having the blog section, the mailing list, and seeing people get hired before there was ever a real platform. It worked.

The Lean Startup Principle Behind TokyoDev's Origin

About the Speakers

— ZooKeep Marketing