Deploying Your First Cove
A Cove is your institution’s private, AI-ready knowledge space. Getting started takes three things: upload a document, let it process, and ask a question. This guide walks the whole loop.
1. Upload a document
From your dashboard, open the Upload panel and drag a file onto it (or click to browse). Thought Cove accepts PDF, Word (DOCX), Excel (XLSX), CSV, plain text (TXT), and Markdown (MD) files, up to 2GB each.
Give the document a clear name and a document type (for example, Academic Catalog or Policy Manual), and set the year or term it applies to. These make it far easier to find and filter later.
Screenshot — The upload panel with a file selected, plus the document name, type, and year/term fields.
Tip
You don’t need to convert anything first. Upload the original file — the pipeline handles parsing, including tables and multi-column layouts.
2. Let it process
Ingestion runs in the background. The pipeline reads each page, analyzes the layout, splits the text into meaning-preserving chunks, and generates embeddings so the content becomes searchable. A large catalog can take a few minutes.
You don’t have to sit and watch — we email you the moment your document is ready to query. You can also check progress any time in your document list.
Screenshot — The document list showing a document in the “processing” state with a progress indicator.
3. Ask your first question
Open the query view, select the document you just uploaded, and ask a plain-language question — for example, “What is the academic probation policy?”
The answer comes back grounded in your document, with inline citations like [1] [2] that point to the exact source passages and page numbers. If the answer isn’t in the document, Thought Cove tells you so instead of guessing.
Screenshot — A question and its answer, with numbered citations and a source panel.
What’s next
Upload a few more documents and try asking a question across several at once. See Asking Good Questions for how to get the sharpest answers.