Candu's AI helps you go from “I need to build something” to “Done and live”. Whether you’re starting from scratch or adjusting something that already exists, Candu AI can draft or refine content based on your prompts and any references you have, including from attachments, webpages or screenshots.
What Candu AI can do and common use cases
Candu AI is especially helpful when you don’t know where to start or you need to make quick changes to existing content. Some examples of what you can use it for:
Making layout suggestions (such as reordering blocks or adjusting alignment)
Localizing copy and images into other language
Creating onboarding checklists
Generating new images or illustrations based on prompts
Designing in-app content from reference documents and images
Tweaking existing content
Create modals for product launches
Draft “what’s new” updates from release notes
This article explains how to use Candu AI, where to find it, and how to get the best results. Let’s dive in!
Get Started
There are two places where you can find Candu AI.
On your Dashboard
You’ll see a prompt: “What would you like to create?”
You can:
Choose from use cases like onboarding users, launching a feature, driving upgrades, or welcoming new users.
Or type your own idea.
Click Generate content to get a draft that you can refine.
Inside the editor
In the floating bar to the right, click the ✨ icon.
Here, you can:
Describe what you want to build
Upload reference files such as PDFs or screenshots
Click Send to receive layout or content suggestions
This is useful for both starting fresh and improving existing content.
Data handling and privacy note: only the prompt and document content you submit is processed. User data is not shared. Uploaded files are handled security and your data is not used to train public models. For full details, see our Data Protection Agreement.
Tips for better results
Use reference files or public URLs when possible. These can include: PDFs, screenshots, help articles, etc. This helps Candu AI understand your context and generate more relevant UX suggestions.
Be clear and specific in your prompt. Vague prompts lead to vague results. Treat Candu AI like a teammate: The more context you sure, the better the output.
Candu AI doesn’t switch between content types. If you’re planning to create a modal, banner, or inline block, start with the correct type before generating content.
What Candu AI won’t do (yet)
Candu AI cannot create forms, flows, video embeds, or conditional blocks yet. It also does not:
Change content types, for example updating an inline onboarding checklist to a tour
Add new tour steps or hotspots [Coming soon]
Update or refine placements or selectors in a snapshot [Coming soon]
Access your content analytics or your user segments
Publish content without your approval
If you run into issues or want to share feedback, reach out to us via chat or email us at [email protected].