Jev could save hours for professors and TAs
User says Jev could save many hours while improving the lives of professors and TAs.
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
User says Jev could save many hours while improving the lives of professors and TAs.
The author shares ongoing work on improving story generation with a Jev-based teacher, pushing the limits of small models and observing student mistakes and questions to guide better writing, suggesting a new teacher-guided approach.
A user shares that Jev automatically answered questions and passed the Alibaba Cloud AI Engineer mock exam in 21 seconds, contrasting with reading and answering each question manually before, highlighting Jev's automation in exam scenarios.
In Jev's Playground, questions commonly asked by Japanese elementary students were answered.
The author shares using Jev to build personalized education courses, noting lesson generation is only ~5% of content cost, with the main bottleneck being validating generated content quality.
Introduces the recent buzz around judgment-focused AI 'Jev', explores its potential use in education for free-text classification and real-time learning support, and notes caution regarding data of students under 18.
The author built a small prototype with TypeSafe AI Jev and Dify to classify preschool teachers' verbal cues, aiming to help teachers reflect on their language use rather than score them.
The author uses JEV to build directed DAGs of science and math topics for grades 6-12, with node and concept-grain prerequisites weighted by JEV noul and strength, tagged as hard.
The author says they don't quite get Jev yet but are curious, and invites others to a Jev hackathon to solve the problem of school handouts crumpling at the bottom of a backpack by tinkering together.
A community member shares an experiment built with Jev and Google Chirp-3 for kids, inspired by another community project. Link
The author shares a school education app built with Jev that classifies class opinions in real time, showing results after inputting 35 people's opinions at once.
Robert Penner posts that Jev AI is a non-verbal savant that wastes no circuits on social skills and focuses solely on beating the Scantron, tagging @typesafeai.