Finished the BlackRock loop a few weeks ago for a Senior SWE role (they call it VP-level internally, which threw me off at first). Sharing the system design round in detail because there was basically nothing useful on the internet about it.
The system design interview was 60 minutes, one interviewer. No virtual whiteboard, just a shared Google Doc. Felt more like a conversation than a drill.
The prompt: design a real-time portfolio risk calculation service. Inputs stream from market data feeds; outputs need to fan out to multiple downstream consumers (trading systems, dashboards, compliance). Not a toy problem.
What they actually cared about: How you handle bursty data ingestion at market open. I talked through Kafka partitioning and a backpressure strategy; the interviewer pushed on what happens if the consumer falls behind. Latency vs. consistency tradeoffs. BlackRock runs Aladdin, so they have legitimate opinions here. I dropped that I understood the domain a bit and it visibly relaxed the room. Schema evolution. How do you update a streaming contract without breaking downstream consumers? I said Avro with a registry, they seemed satisfied. Failure modes. What if the market data feed goes silent? How does the system distinguish "no trades" from "feed is down"?
They did NOT ask the classic "design Twitter" or "design YouTube" questions. This was domain-specific from the start. If you're prepping for a senior or staff-level BlackRock system design round, ditch the generic SDE prep and actually think about financial data pipelines: event sourcing, temporal data, regulatory audit trails.
One thing that surprised me: they asked about observability pretty early. Not just "add Datadog," but what metrics matter and what an oncall runbook looks like. Very ops-aware culture for an asset management firm.
The debrief took about a week. I passed to final rounds. Happy to answer questions about the other rounds too if useful.