Just wrapped a senior SWE loop at Squarespace, NYC remote-eligible role. Posting notes because I couldn't find anything recent when I was prepping.
The system design round is 60 minutes, one interviewer. Format is pretty standard: you get a loose problem statement and you're expected to drive it. The one I got was roughly "design a content scheduling system for a publishing platform" which, given what Squarespace does, tracks. They wanted me to cover: API design (REST vs something else, they weren't dogmatic) Data model for storing scheduled content with versioning How you'd handle the scheduler layer, at-least-once delivery concerns, failure cases Rough scale estimates (they gave me numbers when I asked)
What they cared about: structuring the problem yourself, catching the ambiguities before they had to point them out, and being explicit about tradeoffs. I said "I'm going with Postgres for the job queue over a dedicated queue because the scale here doesn't justify the ops overhead" and they nodded. That kind of thing.
What surprised me: it wasn't a pure distributed systems grind. Squarespace is not a FAANG-scale infra shop. The interviewer seemed more interested in whether I could design something that a team of 8 engineers could actually build and maintain, not whether I knew every Kafka partition strategy by heart. There was a product intuition layer woven in: "how would you change this if the customer could see a preview before publishing?"
Prep I found useful: the usual Grokking the System Design Interview material, but filtered toward mid-scale B2C SaaS, not Google-scale. Knowing basic tradeoffs between relational and queue-backed designs was enough. Staff-level Kafka trivia did not come up.
The level for this role was listed as senior (their equivalent of L5 at FAANG companies, though they don't use those labels publicly). Total round count for the full loop was five rounds including this one.
Happy to answer questions on what I remember.