Careers

Almost everything we build sits between other systems, which means somebody else's operations stop when it breaks.

That might be a partner channel carrying a distributor's whole order book, or a clinical feed where a wrong record does more damage than no record at all. Sometimes it is a gateway sitting in front of a core system nobody is allowed to touch. The constraint is the same either way: the thing has to be right, and it has to keep being right at three in the morning.

That shapes the job. You will spend as much time on failure modes, replay and observability as on the path where everything works. The upside is that what you build tends to stay in production for years, which is rarer than it should be.

Teams are small. Whoever designs an interface usually builds it, and stays on the account after go-live instead of handing it to a support queue. You will talk to the client directly, and you will hear about it directly when something is wrong.