AI Document Processing
When you upload a document, Valiro’s AI reads it, breaks it into chunks, and creates a searchable index. This is what makes semantic search and AI-powered answers possible.
What happens when you upload
Section titled “What happens when you upload”- Text extraction — Valiro reads the text from your PDF or Word document
- Chunking — Long documents get split into logical sections (roughly paragraph-sized pieces)
- Embedding — Each chunk gets converted into a mathematical representation that captures its meaning
- Indexing — Everything gets stored in a way that makes it instantly searchable
The whole process usually takes 10-60 seconds depending on the document size. You don’t need to wait—it happens in the background.
Why chunking matters
Section titled “Why chunking matters”AI works better with focused pieces of text than with entire 50-page documents. When you ask a question, Valiro finds the specific chunks that are relevant and uses those to generate an answer.
This is also why you get precise citations. Instead of “somewhere in this 30-page contract,” you get a link to the exact section.
What processes well
Section titled “What processes well”Best results:
- PDFs with selectable text (not scanned images)
- Word documents (.docx)
- Plain text files
Works but limited:
- Scanned PDFs (if they have text layers)
- Older .doc files
Stored but not AI-searchable:
- Excel spreadsheets
- PowerPoint presentations
- Images
- Audio and video
The AI needs text to work with. If your PDF is essentially a picture of a document (like a scan without OCR), the AI can’t read it.
Processing status
Section titled “Processing status”In the document list, you’ll see status indicators:
- Processing — AI is working on it
- Complete — ready to search and use
- Failed — something went wrong
If processing fails, it’s usually because the file is password-protected, corrupted, or in an unsupported format. Try re-uploading or converting to a standard PDF.
The result
Section titled “The result”Once processed, your document becomes:
- Semantically searchable — find it by meaning, not just keywords
- AI-accessible — Gems can answer questions about it
- Citable — AI responses link back to the source
This is what makes “ask a question about your documents” actually work. The AI isn’t guessing—it’s reading your actual files and pulling real information.
Related: Uploading Documents for getting files into the system. Using Gems for asking AI about your documents.