Convert traffic, measured throughput, and headroom into a small capacity and rollback sheet.
Scenario
A model endpoint must support a peak without hiding tail latency or assuming every replica can run at 100% utilization.
You will demonstrate
Project evidence
Each response is stored in this browser as you type. Include metrics, test output, or a decision rationale wherever the deliverable asks for it.
Record peak traffic, per-replica tested throughput, headroom, replicas, latency SLO, overload behavior, and rollback trigger.
Required evidence: A capacity table with explicit units and one falsifiable go/no-go decision.
0/80 minimum characters
Runnable Python lab
Round up peak demand using only the safe fraction of measured capacity.
Project defense
Rate the evidence, not your effort: 0 missing, 1 weak, 2 adequate, 3 strong. All criteria must be reviewed, but a low honest score does not get hidden.
Capacity, tail latency, headroom, overload behavior, and rollback are explicit.
Completion is controlled by stored evidence, deterministic tests, a decision defense, rubric review, and the artifact when required.
Optional cloud portfolio
An account is required. Submit only when the local completion gate passes. AI review is advisory and separate from deterministic completion.