Determinism and Flakiness in Agent Test Suites
LLM nondeterminism breaks the assumptions underlying traditional software testing.
Jonah Voss
Features Editor
Jonah Voss is a features editor at AgentOps Dispatch covering agent evaluation. Based in San Francisco, Jonah has written for AgentOps Dispatch since 2017.
1 story · San Francisco
LLM nondeterminism breaks the assumptions underlying traditional software testing.