went through PayPal's senior SWE loop in early 2026, specifically targeting an L5 role on their payments infrastructure team. the system design round was the part I was most unsure about going in, so sharing what actually happened.
the prompt was pretty classic fintech: design a payment processing system that handles high concurrency, failures gracefully, and can support multiple payment methods (cards, ACH, wallets). 60 minutes, one interviewer, collaborative whiteboard-style.
what they cared about:
reliability over raw scale. unlike some big tech loops where you immediately jump to "and then we shard to 1000 nodes," PayPal's interviewer kept pulling me back to: what happens when a payment gateway times out? what happens when the message queue goes down? they wanted to see idempotency keys, retry logic with exponential backoff, and a clear mental model of exactly-once vs. at-least-once semantics.
consistency models. not in an academic way, more in a "payments can't lose money" way. we spent a solid 15 minutes just on whether to use optimistic vs. pessimistic locking for balance updates, and what the tradeoffs were for their scale.
the failure path. they explicitly asked me to walk through what a failed transaction looks like end-to-end. dead letter queues, alerting, compensating transactions. this is where I think a lot of candidates underperform. you need to know your distributed systems failure modes.
no "design Twitter" vibes here. everything was payment-context-specific. if you're coming from an infra background with fintech experience, you'll feel at home. if you've only practiced generic LLD/HLD leetcode-style problems, carve out time to think through real payment flows.
round lasted the full 60. interviewer was engaged, asked good follow-ups. overall felt like a real conversation, not a gotcha.
L5 base I was quoted was in the $175-195k range, San Jose. total comp was higher with RSUs but PayPal's equity refresh schedule is worth asking about specifically.