THE JEV GUIDE · CURATED EDITION
Launch news, explanations, demos, and community projects around TypeSafe Jev, organized by topic. Every entry links back to its X source.
Curated selection · Not exhaustive · Updated manually
01 / EXPLORE
0 means awaiting a verified post
Turns a single support ticket into four typed decisions (route, priority, risk, next action) with probability, latency, provider and cost, with an MIT-licensed repo. Link
Using the example of a game support case where the same user sends daily complaints about a character, the post discusses Jev's judgment axes and contrasts it with a human classifying them as fan letters.
Peter Ramsing shares how his team is starting to use TypeSafe's Jev to prompt follow-up questions as tickets are created, noting it must be fast since tickets are often made while on the phone.
The author shares that introducing Jev into a Vane Data-built after-sales support audio pipeline significantly improved judgment accuracy, and notes Vane now supports vane.ai.jev(...) and Relation.jev(...). Link
A remixable demo where typed or spoken feedback is classified and filed into an inbox by Jev from TypeSafe AI, with a cookbook to adapt it.
The author built a real-time chat moderator with Jev that detects thousands of variations and spellings of bad words across multiple languages, noting its extremely low price.
Sam Siam shares Maksym's demo where Jev highlights the next click based on a described need, and imagines it guiding a first-time CRM user through importing contacts.
The author tried implementing AI customer support with Jev for NHK subscription contract questions, using chat to determine whether the user should continue, take action, be exempted, or cancel; Jev handles selecting the next question based on user answers.
Using CowAgent to build a batch ticket analysis tool on Jev: one call per ticket returns 7 classifications, scores, and boolean judgments with probability distributions in about 500ms.
A demo of a support ticket triage tool built with Jev that determines team assignment, urgency, severity, and customer mood in a single call.
The author tries TypeSafe's judgment-focused AI Jev in customer support, noting its speed and accuracy, but warns that even near-perfect results include must-not-miss cases, so role division among Code, LLM, and Human matters.
Routing 72 synthetic support tickets: Jev scored 93% for department, 85% for urgency, and 99% for refund detection; Laya scored 58%, 63%, and 74%, recognizing only 1 of 12 safety tickets.