Tutorial · 4 min

How to use Project Memory (RAG)

A private vector store over your own library

Index your papers and notes into a persistent project memory so every service answers from your own corpus instead of starting cold each time.

Before you start

  • A signed-in SciExpert account

Steps

  1. 1

    Open Project Memory (RAG)

    From the workspace sidebar choose Project Memory (RAG). On a phone, open the All menu and tap Project Memory (RAG).

  2. 2

    Set up your inputs

    Index your papers and notes into a persistent project memory so every service answers from your own corpus instead of starting cold each time. Start with the smallest realistic input so you can check the output before scaling up.

  3. 3

    Run it

    Run the tool and check the first output: per-project indexing of library papers and pasted text.

  4. 4

    Review what came back

    Work through the results: embeddings stored privately to your account; retrieved passages shown with their source; reused by the agent, writer and evidence tools.

  5. 5

    Save or export

    Save the configuration as a preset if you will repeat it, and export the output when you need it outside SciExpert. Your inputs persist across refreshes and sign-ins.

Tips

  • Start with a small input, verify the output, then scale up.
  • Save a preset once the configuration works so you can reuse it across papers.

Troubleshooting

Nothing happens when I run it

Check that every required field is filled and that your plan includes this service — locked services show a padlock in the sidebar.

The output is not what I expected

Reduce the input to a single item, confirm the result is correct, then add the rest back gradually.

More tutorials

Browse all 79 tutorials →