I came back from a career break about a year ago and went through Replit's full loop earlier this year. The behavioral round was something I spent more time prepping than usual because I had the gap to explain, so I have decent notes.
Replit's behavioral questions felt less like a standard STAR checklist and more like they were trying to understand how you think about craft and autonomy. Which makes sense. Small team, high ownership, they're building developer tools so engineers are essentially the target user.
Questions I got or heard about from others who interviewed around the same time: "Tell me about a time you disagreed with a product decision and what you did about it." "Describe a project where you had to figure out what to build, not just how to build it." "How do you decide what not to do when you're resource-constrained?" "Tell me about a time you made a technical decision that turned out to be wrong. What happened next?"
That last one is a values question in disguise. They want to know if you own mistakes and learn, not if you are perfect.
For the gap question: I was direct about it. I said I took time for caregiving and that re-entering the market meant deliberately choosing companies where technical craft mattered more than internal politics. They didn't flinch.
The biggest thing I noticed: they weren't impressed by scale-bragging. "I built X for 10 million users" didn't land the way it might elsewhere. What resonated was specificity about tradeoffs and reasoning. Why you built it the way you did, what you'd do differently.
One round, about 45 minutes, same interviewer who did my final technical.