New Adventures

I’ve been writing this blog (infrequently and inconsistently) for many years now. I rarely (perhaps never) write about my actual work – not due to disinterest, but rather I’m weary of over-sharing. This post is an exception.

A personal career announcement: after 8 years at Riot Games in Los Angeles and Hong Kong, I’ve recently taken a role at the fledgling Supercell Shanghai dev studio. Below I’ll try to recap my experience at Riot as well as look ahead at the new role.

“Thanks Rito”

The Rito meme aside, I do sincerely owe a ton of gratitude to Riot for the various opportunities over these 8 years. Bear with me (or just skip ahead) as I take a long trip down memory lane…

Where it all begun

The story starts around March 2011 – GDC week. I was in my last semester at Berkeley’s MBA program. As anyone with some awareness of MBA programs would know, by March / April, all of the “best” jobs – the Facebook/Apple/Amazon/Google jobs, the consulting & banking jobs – had already finished their formal campus recruiting. The students still searching jobs were either going for unconventional careers (non-profit, startups, entrepreneurship), or just have been quite unlucky/unsuccessful and are probably adjusting their job search strategies.

In my case, I had firmly rejected the idea of going back to consulting.¹ I was also unsuccessful and frankly not passionate/prepared enough for my Google / Amazon interviews. I loved Apple’s products but didn’t see a role that fit me. I had also decided to turn down a potential return offer to VMWare, where I spent the summer – I had a extremely supportive boss, but I knew in my heart that enterprise software wasn’t where I wanted to be.

The only area where I seemed to have some good traction was video-games. Zynga was a super-hot company at that time, and was aggressively recruiting MBA students.² The interview process went fairly smoothly, until the onsite, where I felt I flunked one of the math-heavy analytical interviews (it involved a lot of probabilities / statistics). But this process gave me a lot of confidence that this was a sector that I was passionate about and could be successful in.

Around this time, another serendipitous thing happened, and that was my room-mate Jon started to play this janky Dota-like game (it was League of Legends). I was initially aghast at the art style and graphics; I hadn’t played Dota since 2006, but it was still quite burnt into my memory (especially since I almost failed to graduate college thanks to some Dota binging).

This is all a very long setup to say, when I saw a last minute campus meet and greet event hosted by Riot Games, I almost decided to skip it for an afternoon nap. (I’m very thankful I didn’t.)

The event was hosted by Tom Cadwell, who still leads design at Riot today, and Sean Bender (who went on to co-found Phoenix Labs). It was a very casual chat, with about a dozen students in attendance. They talked about the early trajectory of League; they were also specifically interested in people with backgrounds from Asian markets given their global expansion. That’s me!

From there I was invited to Riot’s GDC party that night, where I spoke with a few more Rioters. Then a whirlwind of interviews, beginning in that GDC week in San Francisco, ending a few weeks later in a half day in Culver City. During the onsite – I had already left the building, before getting called back for an additional round with one of the co-founders. 3 days later I received an offer call.

As an aside – I was also applying to, and interviewing with other games companies. I found more luck with startups. The traditional large companies, e.g. Blizzard and Perfect World, which I remember having some conversations with at their GDC booths, generally lost interest once they heard I hadn’t worked in the industry before.

2011-2014: crazy global expansion

My first role at Riot was as an International Publishing Manager. To be honest, roles at Riot were quite dynamic back then, and I think the key people involved in hiring me also debated a bit about whether I should report to Product or International Publishing.

I didn’t mind either way. My initial assessment was this job was primarily about onboarding myself into the industry, and learning as much as possible. I had generally low expectations of high financial returns at Riot, given the company had just a few months ago had a major transaction with Tencent (which became the majority shareholder).

The first 6 months were a blur. I just remember everything was on fire, all the time – our EU server was overloaded, and our only feasible solution, given the tech we had at the time, was to do a messy server split. Meanwhile, in my part of the org, I was trying to help our regional partners in Taiwan set up their servers and launch the game. The problem was, the thin team supporting these infra buildouts were constantly being pulled in several directions, with for example Korea also eagerly prepping for launch.

The Taiwan launch was a big success – too successful, in fact, as we also quickly overloaded our servers. A peculiar downward spiral in degraded service occurred – to avoid long waits in queues in the evening peak hours, players would log on in the afternoon; during the evening rush hour (where there was a 2+ hour queue), many players that already logged on were idling (went to dinner), leading to long matchmaking times and poor game experience.

It also didn’t help that our deployment tech was extremely shoddy at the time. Deploys, when they went bad, often went over a dozen hours. I witnessed a few of these all-nighters. As I’m not an engineer, my only contribution was making instant cup noodles at 5am to try to comfort the poor engineer still triaging a problem.

In hindsight, what is still amazing is how Riot was able to live through all that. Fundamentally it speaks to the power of the product and the uniqueness of the games-as-a-service model back then – players put up with a lot of really bad instability, and we got better at deploys (painfully slowly, it felt like), and our server stability and capacity grew and grew. In contrast, as many games are operated as live services now, and there are many teams capable of running them, players have more options and thus the service quality demanded is higher.

Around late 2012/early 2013, having helped grow the International Publishing team a little (and the team supported the launch of a few more Southeast Asian markets), I made a lateral career move to a R&D project, nominally to help plan out this product’s international launch plan. I say nominally, as I was much more interested in getting closer to core game development itself, and the main angle I positioned myself was a passion and interest in mobile. I was that guy on the project that was always preaching about the mobile platform.

2014 – 2016 Hong Kong excursion

Late 2013, I got approached for a unique opportunity: move to Hong Kong to help the company jump-start its first office serving China. China had already become Riot’s biggest market by player-base, thanks to the strong publishing of Tencent; however, we felt there were many low hanging fruits to improve and localize the product experience, and we also wanted to strengthen our direct engagement with this massive community.

The choice of Hong Kong as a location was a peculiar compromise. In hindsight I feel quite mixed about this choice. The main rationale were Hong Kong was business friendly (for setting up a foreign-owned entity), geographically closer to our Tencent partners in Shenzhen, and more convenient for western expats. (The main alternative I believe was Shanghai.) However, the critical drawbacks for Hong Kong in hindsight were: very high cost of living, no local game dev industry (tough to hire), and culturally disconnected from Mainland China.

On the career side, this was another lateral move. I officially transitioned into the product management department (“production discipline” as it is called at Riot), and worked on local enhancement features on top of the main League of Legends game.

The big project I worked on was a PC-cafe value-added-services feature, where participating PC-cafes would pay Riot (based on time consumption) for additional content unlocked in League on its machines (for example, all champions unlocked). This was in effect a port of the prevalent business model in Korea, where a substantial portion of a PC game’s revenue could come from PC-bangs instead of directly from the consumer.

This was a challenging project, not really for the tech (by the end the implementation was quite bare-bones), but for the complicated relationships and parties involved, as well as the customer education needed. We were working with multiple teams within Tencent, which meant agendas needed to be aligned. Tencent, as the publisher, also decided to leverage one of the major local telcos as the reseller for this service, which was another compromised choice in my view. (I don’t think a large, state-owned telco’s ground sales force is a good partner to sell an unproven service which needed good salesmanship and hungry hustle.) The results were mixed. Over the course of a year, I believe the service was sold into over 10,000 PC cafes, but that’s out of a universe of over 100,000 PC cafes in China. And we saw a muted player response.

Fast forward to end of 2015, and a bunch of things that had been brewing for a while led to Tencent purchasing the remainder of Riot. This was a significant equity event, even for folks like me who joined after the 2011 majority purchase.³

At this time, I was also again at a cross-roads. On the personal front, our first son was born early 2016, and my wife and I knew we weren’t going to stay in Hong Kong much longer – we just thought it was too cramped to raise a kid there. Professionally, the newly opened Riot Shanghai office offered opportunities to go deeper on the China market, while in Los Angeles there was a fledgling R&D-ish project I could lead.

2016-2019 adventures in Riot R&D

We decided to come back to LA, and we made that move at the end of 2016.

Career-wise this was yet another lateral move. In some ways, I have been very bad at climbing the Riot corporate ladder (and yes, as a company with over 2,000 employees at this point, there was definitely a long ladder to climb if you wanted to). I enjoyed jumping from team to team, and I passed on some advancement opportunities (e.g. stepping up to lead a team in an earlier role) in favor of exploring new areas via lateral moves. I would be dishonest if I said I was perfectly happy with these outcomes – because of course I do care about the size of my paycheck and the leadership responsibilities I have – but overall I’d probably make the same choices if presented with these opportunities again.

I started a project with two other Rioters, in a workspace converted from a small meeting room in a corner of the campus. Later on as we grew and struggled to find space, we were in the literal basement for a few months. This was an extremely uncommon project – it was outside the purview of the formal R&D organization, tucked in an organization that many Rioters in LA would probably struggle to define. Unfortunately I can’t go into specifics – but I’m extremely proud of the windy road we took over the next couple of years, growing the team to a couple dozen of Rioters, and eventually becoming an official R&D project. There are many crazy (good and bad) stories to tell, and hopefully I will have a chance to tell them when this project – which really is like another baby of mine – ships.

The personal stuff

Last but not least in this reminiscence of my Riot days, I should do some accounting of my personal life. Having spent 8 years at Riot, there are so many personal life events intertwined, and I do feel a lot of serendipity and connection.

My wife and I – we first met during my interview trip to Riot in April 2011. Well, technically we already knew of each other (we went to high school together), but it was during that brief trip, where both of us happened to be traveling in LA, where we bonded.

When I decided to move to Hong Kong, my wife begrudgingly accepted and effectively put her career on hold to be with me. And as I said earlier, in early 2016 we had our first child, our son, in Hong Kong.

And also in the spring of 2019, we had our second child, our daughter, in LA.

Perhaps I’m being overly sentimental, but in a way I see my 8 years at Riot also as an unforgettable life journey, where I started it while in the early-blooming of a relationship, and ended it with a happy and fulfilled family (no more kids!). And this is where I have to say, I’m eternally grateful for my wife, who has always supported me in whatever foolish / selfish decision I’ve made (despite not agreeing with them), and who has absolutely carried the lion’s share of the hard work that keeps the family running.

(Oh by the way, the move to Shanghai is another one of these selfish decisions that my wife just fully supports me in.)

Joining Supercell

So finally we come to the new adventure.

In terms of how I got here, it’s a bit of a cliché. Jim Yang, who heads the Supercell Shanghai office, previously worked at Riot and was for a while part of the Hong Kong team. We’ve stayed in touch since then, and ever since I’ve been going deeper into the rabbit hole of games product management,⁴ Jim has been interested in discussing opportunities in Shanghai.

I’ve generally turned down those opportunities, as I had myriad reservations about going back to China, both personally and professionally. But a few things got me over the hump a few months ago.

For one, I had been at Riot a long time. I still found it weird to be called an old-timer, because I saw the pre-2011 buyout Rioters as the real old-timers, but objectively I fit the label. And to be perfectly candid, I don’t naturally gravitate towards working at a large organization, which Riot now definitely was – most of my career I’ve been drifting the other way, towards smaller teams. ⁵

Secondly, I’ve always wanted to work on mobile, and even though Riot now finally takes mobile seriously (and I’d like to think I played a small role in that), many (perhaps most) Rioters’ in their hearts of hearts still live and breathe PC. I have this grave imposter syndrome feeling – that I’ve been a flag-bearer and self-labeled expert for mobile games (especially Chinese ones) at Riot for a long time, but I really lack direct inside-the-trenches experience, especially working alongside mobile veterans.

Thirdly, even on this very blog I’ve often proclaimed the rising tide of Chinese devs and the complicated games they were shipping globally. If I truly believe my vision, I should be trying to ride this momentous tide in my own career.

The role at Supercell Shanghai meets all these points. Without waxing poetic too much – I’ve long admired Supercell as a games studio, with a track record of shipping industry-leading mobile games.⁶ The Shanghai dev studio is a high risk venture, aspiring to make Supercell quality games with a diverse small team of local Chinese and international talent, but the risk is proportional to the size of the opportunity. And I’m joining the studio at an early enough time, with an opportunity to substantially help shape the culture and grow the team.

Perhaps a bit ironically – earlier this year I had written a quick post talking about Supercell’s challenge, in response to our CEO Ilkka Paananen’s open letter. I certainly didn’t think a few months later I’d be interviewing with Ilkka, or that it would be part of my job to try to solve this challenge…

Anyhow, it’s time to wrap up this incredibly verbose and overlong post. I’m sure there are lots of typos and such. But I did feel compelled to write a summary, and for those of you that read it to this point, I hope it was worth your time. Hit me up on twitter (@NanDuan) if you’d like to chat.