Codex case library›Workflows and Agents
Back to all cases

Open-Source AI + Kindergarten FDE: Archiving Growth Materials with a Local Large Model

A dozen or so videosAmount of material teachers send to the group daily
2 weeksValidity period of group-chat material information
WeeklyFrequency of teacher file verification
01

What this case covers

The author open-sources an AI + kindergarten FDE solution covering the pain points of three groups: parents, teachers, and the kindergarten. The article discusses material archiving, growth assessment, and commercialization ideas; full collection by the kindergarten remains only at the proposal stage.

02

Questions it answers

  1. Why can't parents see their own child's materials?
  2. Why can the solution only use a local large model?
  3. How is the information teachers enter uniformly archived?
  4. Why doesn't many-to-one communication work for now?
  5. Where is the idea of full collection by the kindergarten stuck?
03

Our observations

  1. What parents want is their own child's portion, not the whole class's collection.
  2. Compliance for sensitive information is a hard constraint of local deployment.
  3. The full-collection idea is limited by storage and token costs.
04

From the author

Only if collection itself does not discriminate by time or by child can the later analysis possibly be complete.
QR code for a three-day trial of Shengcaiyoushu

The complete post is available in the Shengcaiyoushu community

Scan for a free three-day trial and read the complete original. Platform rules apply.