Storage
In the previous lesson you created tables and the endpoints for your app. Here you ask your AI assistant to implement Storage in DYPAI: it will check whether buckets already exist, create them if needed, and wire everything so tasks can have file attachments.
What this video covers
Prompting the AI
You send a prompt describing what you want (tasks with attachable files, etc.). The assistant looks for existing buckets, checks credentials, and implements the flow.
Model quality
The model you use does not have to be the most powerful on the market. DYPAI works well with a wide range of models. Stronger models tend to produce fewer mistakes and cleaner designs, but they are also more expensive. Smarter models generally mean better backend output; the important part is that DYPAI fits your workflow regardless of tier.
If something breaks
If a small error appears, you can tell the agent again—it can correct itself automatically, then you retest.
Testing uploads
You upload a test file, save, and DYPAI stores it in the buckets: files live under the user scope, you can list them, download, and delete them. The whole flow stays simple compared to what file handling often feels like elsewhere.
What you'll learn
- Having the AI create or reuse buckets and implement attach files to tasks
- How model choice affects quality and cost
- Iterating with the agent when something fails
- Upload, list, download, and delete files in Storage
Next
Continue to AI agents & chatbot—where you will build an AI assistant with tools so it can do more than chat: for example create tasks for you, not only answer questions.