Compressed caveman-style prose for AI coding agents — cuts ~65% output tokens while keeping full technical accuracy
96
100%
Does it follow best practices?
Impact
96%
1.00xAverage score across 38 eval scenarios
Passed
No known issues
{
"context": "Tests whether the response correctly maps scenarios to load balancing algorithms.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Stateless API: Round Robin",
"description": "Recommends Round Robin (or Random) for identical stateless servers — simplest and sufficient",
"max_score": 12
},
{
"name": "Mixed servers: Weighted or Least Connections",
"description": "Recommends Weighted Round Robin, Weighted Least Connections, or Least Connections for heterogeneous server sizes",
"max_score": 12
},
{
"name": "Chat app: Sticky Sessions / IP Hash",
"description": "Recommends sticky sessions (session affinity), IP hash, or consistent hashing for the WebSocket use case",
"max_score": 12
},
{
"name": "Explains why each algorithm fits",
"description": "Provides reasoning for each choice, not just the algorithm name",
"max_score": 10
},
{
"name": "No incorrect information",
"description": "Load balancing algorithm descriptions and behaviors are technically correct",
"max_score": 10
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17
scenario-18
scenario-19
scenario-20
scenario-21
scenario-22
scenario-23
scenario-24
scenario-25
scenario-26
scenario-27
scenario-28
scenario-29
scenario-30
scenario-31
scenario-32
scenario-33
scenario-34
scenario-35
scenario-36
scenario-37