Uploading Documents

Add PDFs, Word documents, and text files to your Sources for AI-powered retrieval.

Documents are one of the most common ways to populate Sources. Upload your existing documentation, policies, guides, and manuals to make them searchable by AI.

Supported File Types

PDF

Most common format for business documents

Word Documents

Microsoft Word files (.docx, .doc)

Text Files

Plain text documents (.txt)

Uploading Documents

1

Open Your Source

Navigate to the Source where you want to add documents.

2

Go to Documents Tab

Click the "Documents" tab at the top of the Source page.

3

Click Upload Document

Click the "Upload Document" button.

Documents tab with Upload Document button highlighted
Click Upload Document to add files to your Source
4

Select Files

Choose one or more files from your computer. You can also drag and drop files.

5

Wait for Processing

Hyperleap processes your documents, extracting text and creating searchable chunks. This may take a few moments for large files.

Document Processing

When you upload a document, Hyperleap:

  1. Extracts text — Reads all text content from the file
  2. Chunks content — Splits into searchable segments
  3. Creates embeddings — Generates vector representations for search
  4. Indexes — Makes content available for retrieval
Tip:
For best results, use well-formatted documents with clear headings and structure. The AI retrieves relevant chunks, so organized content improves accuracy.

Document Status

After uploading, documents show their status:

Processing

Document is being analyzed

Ready

Document is indexed and searchable

Failed

Processing error (check file format)

Best Practices

File Quality

  • Use text-based PDFs — Scanned images need OCR processing
  • Check formatting — Tables and complex layouts may not extract perfectly
  • Remove password protection — Encrypted files can't be processed

Content Organization

  • Use clear headings — Helps chunk content meaningfully
  • Keep documents focused — One topic per document works better than huge combined files
  • Update regularly — Replace outdated documents with new versions
Note:
Large files (50+ pages) may take longer to process. For very large documents, consider splitting into smaller, topic-focused files.

Managing Documents

In the Documents tab, you can:

View

See document details and extracted content

Download

Re-download the original file

Delete

Remove a document from the Source

Common Issues

Text Not Extracted

If your PDF shows 0 pages extracted, it may be:

  • A scanned image without OCR
  • Password protected
  • Corrupted file

Poor Retrieval Quality

If AI responses don't reference your documents well:

  • Check that the Source is connected to your chatbot/tool
  • Verify documents are in "Ready" status
  • Try asking questions using terms that appear in the documents

Next Steps

Learn how to add website URLs to your Sources for indexing web pages.