Docs & Data Lab

Docs & Data Lab

This is the reading room. Feed it PDFs, Word files and tables, see what structure the machine finds and build a tiny search index so you can question your own documents.

Security reminderThis studio is for education and experimentation. Do not upload production data or secrets. Outputs are demos; review before using anywhere safety-critical or financial.

1. Upload and inspect documents

Free tier limit: 4.0 MB per file. With credits: 12.0 MB. Allowed: pdf, docx, txt, csv. Do not upload real customer data or secrets; use synthetic or anonymised samples.

Templates and spec

Download small examples you can upload immediately, plus a simple spec of what the lab accepts.

Download example CSV (small)Download spec (JSON)Data prep mini dataset (CSV)
  • Keep uploads under the per-file size limit shown above.
  • CSV should include a header row.
  • Use synthetic or anonymised data only.
No documents yet. Upload a PDF, DOCX, TXT or small CSV.

2. Text and entity explorer

Upload and select a document first.

3. Build a retrieval index

Select documents, build a tiny index, then ask questions against your own content.

Upload at least one document first.

4. Schema and field mapper

Paste two sets of headers or keys. We will suggest simple matches to show how schema alignment feels in integration work.

Source schema

Target schema

Suggested mappings

Add both schemas to see suggestions.

    How to use this

    This is a sketch of schema alignment. Adjust field names and confirm mappings manually. Unmapped fields highlight gaps to address before integration.

    5. Recent docs runs

    Latest index builds and document actions. Open Control Room for the full view.

    Open Control Room
    No runs yet. Build your first index to see it here.