Can Jev choose which tool an agent calls?

Yes: TypeSafe’s own function-calling cookbook has Jev answer one choice question, “What is the user asking the trading assistant to do?”, with every tool’s description from spec.json listed alongside the built-in answers, and it returns a confidence with whichever function it picks. That confidence is the call’s weakest link, the least certain of the tool choice and every argument together, not a product, because one wrong argument is enough to spoil the result.

The cookbook’s own fourteen test commands show the range: a clear request like “what tickers do you have” scored 1.00, while an ambiguous one, “when during the day does nvda trade the most,” scored 0.53. The same threshold pattern used for any Jev decision applies here: route the low-confidence picks to an LLM or a person instead of calling whatever tool Jev happened to choose.

sources

keep reading

More on this.

Can Jev replace an LLM judge for grading agent traces?Partly. Jev can take the narrow yes-or-no parts of a judge's rubric, but not a judgment that needs reasoning across a whole trace or a written reason for the verdict.answer →Does Jev beat a frontier LLM's accuracy on the same task?On TypeSafe's own benchmark, Jev averaged 67.8% accuracy at $0.0004 a case, beating Claude Opus 5 run as a single prompt (64.8% at $0.34).answer →How do I test Jev against the LLM call it would replace?Run both against the same held-out labeled sample and compare accuracy and cost per correct decision; one test found splitting the question flips the result.answer →Everything on jevThe definition, and every question answered under jev.concept →

Two ways to run Tessary.

Tessary is an open-source agent reliability platform. Cloud and self-hosted run the same workflow on the OpenTelemetry traces your agent already emits.

Tessary Cloud

We host it for you. Send your first trace with nothing to deploy and no model key.

what's includedper organization
traces
10,000 per calendar month
stored trace data
1 GB
retention
30 days
model credit
$10, one-time, for triage and root-cause analysis
credit card
not required
start for free

Self-hosted Tessary

Run the open-source code on your own infrastructure with one command. Add your own model key for triage and root-cause analysis.

Self-host Tessary for me by following https://github.com/tessaryai/tessary/blob/main/setup.md

docker compose -f oci://docker.io/tessaryai/tessary:compose up -d -y