Description
I'm looking for a backend engineer who's kept a live game up and running while it was actively on fire.
The kind who's been paged at the worst possible moment, found the problem, and shipped the fix without taking the whole thing down in the process.
Not "write the backend and hope staging was honest." Staging is never fully honest.
The real test is production, at peak, mid-event, with real players and real money moving through it.
We're building and running a live mobile game in Tokyo! A worldwide IP, growing fast, the kind of system where a bad deploy is felt by a lot of people the moment it lands.
I want someone to own big pieces of the backend that keep the whole thing running.
Not a ticket-taker.
An owner.
And this one's rare: you'd work shoulder to shoulder with the people actually making the calls, the founders, the leads, the ones deciding what we build and why.
No layers, no broken telephone.
Your engineering judgment goes straight to the table, where it moves things.
One thing that's not up for discussion: this is full on-site in Tokyo.
Not hybrid.
Not remote.
No exceptions.
You don't have to be in Japan today, but you do have to be genuinely up for relocating here, because we're building something hard and fast, and that only happens with everyone in the same room.
You'd be deep in the real work: backend services in Go, gRPC, protobuf, MySQL, and Redis behind inventory, IAP, live events, leaderboards, guilds, notifications, player state, and live config.
Systems that have to survive event launches, timed campaigns, content updates, rollbacks, partial failures, and clients running three versions behind.
The hard part isn't writing it.
It's deploying updates and firing off events while production is full of live players...
and having it be a complete non-event, because you built it to be.
And when a major event goes live, your job is to make sure the backend doesn't quietly start looking for another career.
I have a specific person in mind.
The kind whose teammates would describe as:
calm at 3am when the graphs go the wrong waythe one who owns the incident instead of hunting for someone to blamethinks in failure modes: retries, idempotency, "what happens when this breaks?"writes the boring, reliable code and the dashboard to watch itclear when everything around them is unclear
What you'd bring (Requirements):
5+ years of backend / server-side engineering on real production systemsYou've designed, built, AND operated complex backend services end-to-end.
Not just written them, run themGo (or you'll pick it up fast as we live in Go, gRPC, protobuf), plus real fluency in at least one more language (Node.js, C#, whatever)Strong with relational databases: data modeling, transactions, consistency, performance (MySQL here)Solid Unix and networking fundamentals! Files, sockets, service-to-service comms, and comfortable with Docker/containersYou reason clearly about reliability, scale, latency, retries, idempotency, and failure modesSharp production debugging instincts: logs, metrics, traces, DB and cache state, service health, real player impactCS degree, or you learned it the hard way — both count
Bonus points!(not mandatory, but great to have for a perfect fit):
You've built live-service game backends before (events, IAP, rewards, leaderboards, progression, social, live config)You've stood watch during launches, traffic spikes, and incidents, and lived to tell the taleAWS, EKS/Kubernetes, cloud-native experienceYou've worked with Unity client teams or know enough game dev to speak their languagefull-stack rangeand a healthy respect for that one suspicious Redis key nobody wants to touch
This isn't a seat for someone who wants a quiet queue of well-defined tickets and a system that never changes.
The system changes constantly, on purpose, while people are playing.
That's the whole job, and if you're the right person, that's the fun.
If that's you, complete every field in the application, even the ones that aren't required.
That's how you stand out.
And if you already know you can carry more than your current stack lets you carry, if you want a system big and alive enough to actually test you, and a team that'll push you to your limit and put the wind at your back the whole way there, apply yesterday!
Otherwise, keep watching the dashboards from a safe distance while the rest of us keep the thing alive and growing.
