Jev in production › Writing and content scoring
Passage (Working Memory Jev)
Experimental localhost teaching aid has Jev make bounded choices about familiarity, grouping and relationships at each reading step to estimate working-memory demand in instructional text; analyzing the sample passage cost $0.050 (author).
- Use
- Writing and content scoring
- Industry
- Education and research
- Form
- Open-source tool
- Stage
- Beta
- Listed
- 2026-09-22
- Found via
- github
- Repository
- AustinAWay/Working-Memory-Jev
- Stars
- 76
- Forks
- 12
- Last push
- 2026-09-21
- Language
- Python
- License
- none stated
The README opens with
Passage helps educators inspect where instructional text may ask a learner to hold too many ideas or relationships at once. It runs on localhost and uses the real Jev API to propose active groups, source connections, and changes in demand as the passage unfolds.
This is an experimental teaching aid. Its slot estimates and confidence labels are not validated measurements of a student's memory or calibrated probabilities of overload. Research motivates the design; it does not validate the app's numerical predictions. See research and practical use.
Badge
For the project's own README, linking back here:
Also used for writing and content scoring
- Jev X Sentiment Analysis ~$0.0008
- Osso about a tenth of a cent
- riff $0.0004
- Proseward 50x
- dotmap.ai about half a cent
- Postneedle about four thousandths of a cent
- Community Signals (Joshua Poddoku) 77%
- Maxfusion $0.00028