← Selected work

Performance engineering practice

Performance at Scale

Load and observability work focused on finding constraints before they become customer incidents.

k6observabilitysystems

The operating condition

Peak behavior cannot be inferred from a passing functional suite.

How the work was structured

Model realistic load, observe the generators and system together, then move p95/p99 and error thresholds into delivery gates.

What can be inspected

  • A documented 100k RPS cloud test scenario
  • Distributed k6 execution and New Relic analysis
  • Smaller CI performance gates for regression detection