Hiring senior engineers across three time zones — what actually works.

Auto-generated and lightly edited. Let us know about errors.

Adaobi Eze: Hiring is an unsolved problem at Spalce in the sense that we're constantly evolving the playbook. Sara, you lead People for the engineering org — what does our team look like today? Sara Mensah: We have just over 120 engineers across three primary hubs — Accra, Nairobi, and London. Plus about 25 fully remote engineers in other African and European cities. The org is structured so that most teams have at least two of the three hubs represented. Adaobi Eze: And that's a deliberate choice. Single-location teams are easier to manage but the talent pool is smaller. Three-hub teams are harder to coordinate but you can hire from a much larger pool of senior people. Sara Mensah: The first lesson — overlap matters more than headcount. We tried for a while to put one engineer in Lagos with the rest of the team in London. The time difference is only one hour but the working day patterns are different. The Lagos engineer was always slightly out of the conversation. We've now made it policy that any new hire joins a team where they have at least four hours of overlap with another team member, daily. Adaobi Eze: The second lesson is about written communication. Distributed engineering is a writing job. If you can't write a clear architecture document, a clear pull request description, a clear incident postmortem — you'll become a coordination tax on everyone around you. So we test for that explicitly. The first stage of our interview is a written technical exercise where we look at clarity and structure, not just correctness. Sara Mensah: We've also evolved how we think about compensation. We used to pay local market rates — London rates in London, Accra rates in Accra. We found that we were losing our best Accra engineers to remote-first international companies that paid London rates regardless of location. We now pay global bands. It costs more on paper, but the retention numbers justify it. Adaobi Eze: A counterintuitive finding — meetings get more important, not less, in distributed teams. We tried very hard to be async-first. The result was that decisions stalled, context decayed, and silence got mistaken for agreement. We now run a weekly sync per team — sixty minutes, agenda required, decisions captured in writing. Plus a monthly all-hands. Sara Mensah: On the hiring funnel itself, we found that our application volume increased significantly when we listed roles as fully remote within Africa and Europe, rather than location-specific. We also found that our senior conversion rate improved when we published the salary band on the job posting. There's still some uncertainty around that — for context, in some markets transparent compensation is unusual — but the data on our side is clear that it improves the quality of the applicant pool. Adaobi Eze: One thing I want to highlight — interview loops should be short and decisive. Our loop is four rounds total: written exercise, technical pair, system design, and culture conversation. We make a decision within ten business days of the first response. Senior engineers have options. If your loop is six weeks, you'll lose them. Sara Mensah: And finally, onboarding is when most distributed-team problems compound. We invest heavily in the first thirty days. A buddy in the same hub, a buddy in a different hub, weekly structured check-ins, and a documented first-thirty-days project. Get that wrong and the hire never feels like part of the team.