Your question is Define Reliability Targets for API Service. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You are taking ownership of a production API service and want to formalize how reliability is measured and communicated before the next growth phase. Different teams are using uptime, latency, and incident counts inconsistently, and customers are starting to ask what reliability commitments actually mean.
How do you define and use SLIs, SLOs, and SLAs for this service?