Eval Library
T
For TemporalCode Assistant

Task Queues Worker Scaling

Temporal · Temporal

Temporal evals — Task Queues & Worker Scaling (relift v3)

About Temporal

Temporal is an AI-powered platform helping teams automate complex workflows, surface insights, and deliver better outcomes through intelligent automation.

Employees

100–2,000

Industry

Enterprise AI

Headquarters

United States

Sample tests· showing 3 of 6

#InputExpected behaviorCheck
01

Team disabled sticky execution to debug; workflow task latency 3x. Docs: sticky queue keeps workflow on same worker for cache.

Agent re-enables sticky workflow task queue settings per SDK defaults, scales workers horizontally, documents tradeoff.

Pass / FailTool usemedium
02

Activity task backlog 50k; each worker maxConcurrentActivityExecutionSize=1 with 2 pollers.

Agent raises poller count and maxConcurrentActivityExecutionSize within CPU/memory bounds, splits hot queues, tags [REQUIRES-VERIFICATION] on Cloud poller limits.

Pass / FailTool usehigh
03

Sales asks guaranteed 50k workflow starts/sec on Cloud shared tier.

Agent cites discovery_gap, recommends load test and Temporal support sizing, tags [REQUIRES-VERIFICATION] on SLA numbers, does not invent RPS cap.

Pass / FailPolicymedium

Rubric criteria

  • Temporal
  • Durable Execution
  • Task Queues Worker Scaling

Recommended for

TemporalTemporal customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.