devwithjev
β€” reading nowβ€” views
Submit a build

Evaluates and Repairs Shorts with Jev

Erico Siu ran four Shorts through Jev in Codex to check whether titles matched transcripts, endings were complete, and essential context was present. Codex showed repairs before edits advanced, and a person reviewed the final pacing and visuals.

View on X
ericosiu@ericosiu𝕏
We ran four Shorts through Jev in Codex. For each cut, it checked whether the transcript supported the title, the thought had a complete ending, and any essential context was missing. When a check failed, Codex showed the repair before the edit moved forward. Cut β†’ evaluate β†’ repair β†’ review. Then we feed the result back into the eval and tighten the criteria for the next run. A person still reviews the final edit for pacing and visuals. I broke down the full workflow in the article here πŸ‘‡
Sep 23, 2026X postsView on X
The results were fed back into the evaluation, and the criteria were tightened for the next run.

Also filed under Content & growth

  • Real-Time Emotion-Driven Character Animation

    The creator used Jev for real-time emotion extraction on a character exported from Tripo P2.0, combining it with Audio2Face lip sync and emotion-based body, eye, and blink performance. The build also includes microsaccades and DLSS 5.

  • X Post Virality Analyzer

    Movez built a viral post prediction analyzer with Jev and Opus 5.5. It compares an X link against 800 viral posts and gives it a virality score.

  • Organizes Posts with Contextual Icons

    Coinathlete is trying to organize posts using JEV and says the icons and labels look good on each post. They made 1,001 individual icons based on classified context.

  • Jev Grades AI Assistant Work

    JH Trader uses Jev with Muse to review AI assistant Sarah’s work before he sees it.