JevMade hello@JevMade.com
← Back to guides

JevMade field notes / Worked recipe

Add searchable labels to product descriptions

This recipe turns product descriptions into categories and tags that a database can search, while leaving prices and exact comparisons to ordinary software.

Original by Jeroen Erne / NexibeoClassification

Listen to this guide

JevMade’s plain-English explanation

0:00 /

AI narration

Credits and license

“02 · Database indexing” by Jeroen Erne / Nexibeo. Read the original source.

This expanded guide is an AI-narrated adaptation of the source’s essential explanation, examples and caveats, not a word-for-word reading. The synthetic voice does not imitate the author or imply their endorsement.

License: MIT

MIT License

Copyright (c) 2026 Jeroen Erne (https://nexibeo.com, https://completeaitraining.com)

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Our summary

A shop's descriptions may suggest that an item is a gift or suits a certain audience without storing those details separately. This recipe asks Jev to identify categories, audiences, and tags, then saves its answers alongside the unchanged product records.

The program checks which products still need labels before sending requests. It stores categories, probabilities, the model used, and the indexing date in separate tables. The database can then combine those labels with exact conditions, such as a maximum price, without asking AI to do arithmetic.

Some labels in the author's sample were debatable or wrong. A saved probability helps someone find uncertain records but does not prove the other labels are correct. Clear tag definitions matter, and changed questions or model versions may require checking and labelling products again.

Key takeaways

  1. Preserve the original records when adding inferred labels.
  2. Store the model, date, and confidence with each judgment.
  3. Use database rules, not AI, for prices and arithmetic.

The recipe uses a small product sample. JevMade has not run the indexing program or verified the author's results.

GitHub cookbook · Source reviewed

Read the original guide Opens the author’s site in a new tab.