Start a grounded task - Nowledge Mem 101

AI Now · Lesson 1 / 5 ~5 min Intermediate

Question

Source

Evidence

Investigate

Save

Start an AI Now task in Nowledge Mem with one question about your own work. Before asking for a recommendation, check which saved facts and constraints it found.

What you’ll accomplish

You’ll start an AI Now task that identifies the saved context relevant to one real question.

Learn with the simulation

On desktop, click Start the simulation under Your turn when you are ready. Follow the popup hints to send a request, read its reply, and check the result. Each lesson prepares its fictional Atlas materials; no model setup is needed.

The simulation runs locally with preset replies. It does not call AI or search the web, and refreshing resets the practice. On mobile, read the steps first and use a desktop browser for the interactive exercise.

Before practising in the real app

Configure and test a tool-capable model for AI Now using the LLM Providers guide. You can use a supported local or hosted provider. Keep the Mem server running. If you connect to a remote Mem server, AI Now uses that server’s model configuration. Choose a real decision with at least one relevant saved memory. To practise with the course example, paste this fictional context into Timeline and press Enter first:

Atlas planning context — fictional practice scenario

Goal: decide whether to hold one cross-team planning review next quarter.
Scope: use existing quarterly notes to check dependencies over the next twelve months.
Not decided here: how often future reviews should happen.
Current practice: each team already runs quarterly reviews.
Constraint: reuse existing review notes; avoid a second reporting process.
Open question: would a broader review reveal cross-team dependencies earlier?
Status: no pilot has been approved.

Your turn

Click Start the simulation and follow the popup hints. After practising, use the steps below to repeat the workflow with your own material in Nowledge Mem.

Open AI Now

In Mem, open AI Now from the sidebar and begin a new conversation. Keep this conversation for lessons 2–5. If you use Spaces, select the one containing your task memories. Otherwise, stay in Default.

Ask a question tied to your knowledge

Use a question with a specific decision and time frame. For example:

Search the saved Atlas planning memories in Nowledge Mem. We are considering one cross-team planning review next quarter, covering dependencies over the next twelve months. We are not deciding a recurring review schedule. List the relevant facts, constraints, and undecided questions, and identify the memories behind them. Use saved knowledge only; do not research the web or recommend a direction yet. Say which context is missing.

Check the context before moving on

Open the memories behind the summary. For the practice example, check that it preserves the twelve-month planning scope, existing quarterly reviews, the constraint against extra reporting, and the unapproved status. It must not turn the planning horizon into a yearly meeting schedule.If a memory is missing, type @, find it, and select it in the conversation. Ask AI Now to revise the context summary. If it cannot find the memory at all, confirm the memory was saved in the current Space.

Start when ready; practice hints appear in the popup.

What you should see

AI Now names the saved context it used and preserves the goal, constraints, and undecided status. Keep this checked summary for the source-reading step.

Why this matters

Starting with grounded context keeps the task connected to your actual history. It gives you a better question to investigate than a broad request for advice.

Go deeper

Read AI Now in the Nowledge Mem Docs for setup and context behavior.

​What you’ll accomplish

​Learn with the simulation

​Before practising in the real app

​Your turn

​What you should see

​Why this matters

​Go deeper

What you’ll accomplish

Learn with the simulation

Before practising in the real app

Your turn

What you should see

Why this matters

Go deeper