A demo of daily AI-made personal summary pages
Balladic showed a demo where each user gets a new Narrative page every night. It reads activity, comments, and changelogs, then turns them into a book-like summary. For solo makers, it is a useful example of AI making a product feel more personal.
Key points
- Each user gets a different Narrative page every day.
- The system summarizes changelogs, comments, and recent activity.
- It uses several open-weights models to create and check the text.
- It also compresses data to reduce token costs.
Quick term guide
- Narrative
- Here, it means a custom summary page that tells the story of recent activity.
- changelogs
- Changelogs are notes that list what changed in a product or project.
- open-weights models
- These are AI models whose core files are available for people to run or adapt.
- open-weight
- The model's internal numbers are publicly released, so anyone can download and run or modify it freely.
- weights
- The internal values an AI model learns during training; sharing them is what makes a model truly runnable by others
- models
- Different AI engines that can power answers or code suggestions inside a tool.
- token costs
- Token costs are the fees paid for the text an AI model reads and writes.
- token cost
- The money or usage spent when sending text to an AI model and getting text back.