What happens when a Tessary classifier flags a trace?

Crossing the threshold writes a finding: a record that one trace tripped one classifier, and the score that did it. Nothing pages anyone yet.

LLM judgment only enters at this point, reading the finding instead of the raw trace, which is also where the model cost in the pipeline actually sits: the bill tracks how often the cheap checks fire, not how much traffic you serve. Triage then separates a real deviation from legitimate change, the harder half of the job. A tool your team changed on purpose looks identical to drift until something checks it against what actually happened.

What survives triage becomes a case, and a case is what gets attributed and, if it clears the alerting rules, what reaches a human. A flag that doesn’t survive triage never does.

keep reading

More on this.

Send us the traces you already emit.