- Workforce role AI Tutor Employee
- Vertical Education
- Agent Syntic Visualization Specialist Agent
- Capabilities Custom interactive visuals
- Share Copy link
The request
A stats student tells the AI Tutor Employee on their workspace: "I'm learning linear regression. I can do the formula but I don't get it. Build me something where I can mess with the data points and watch what happens to the line." The Tutor reads the verbs — mess with, watch, feel — and dispatches a Syntic Visualization Specialist Agent to ship an interactive canvas, not a passive demo.
Within seconds the Agent returns an empty scatter with toggles for residuals and influence. The student places points, drags an outlier, and the squared-distance weighting clicks for the first time.
Prompt the AI Tutor: "I can do the formula but I don't get it. Hand me a canvas where I drag points and feel why one bad value throws the whole fit off."
How the Tutor briefs the Agent
No files needed. The Tutor parses the verbs and sends the Specialist Agent two instructions: build a manipulable surface, expose residual and leverage toggles. That brief is what turns a static diagram into a workbench.
What the Agent needs
Nothing to upload.
The Tutor handles the rest from the brief.
What ships back to chat
The Specialist Agent returns an empty canvas. You click to place points, drag to break things, watch the line respond. Placing the outlier yourself and watching the line pull toward it is what makes squared-distance weighting click. The influence toggle draws a halo around whichever point the fit depends on most — drag one to the edge and the halo grows even when the point sits right on the line.
Shipped by the Syntic Visualization Specialist Agent.
Follow-up dispatches
Keep the conversation going — the Tutor dispatches new Agents as questions sharpen.
Surface the math behind what you noticed
Something you saw while dragging raises a question. The Tutor dispatches a Math Agent that ships a second visual showing the formula in action.
"The influence halo grew when I moved the point sideways but barely changed when I moved it up. Show me the leverage formula and where horizontal distance shows up in it."
Overlay a second model on the same canvas
Ask the Tutor to add a second technique on the canvas you built — both stay live, and dragging shows how they respond differently.
"Overlay a robust regression on the same points I made. I want to drag the outlier and watch both lines move."
Turn the canvas into a prediction test
The Tutor dispatches an Assessment Agent that generates test scatters, asks you to predict, then tells you if your instinct was right.
"Give me a few test patterns — show me a scatter and ask me to predict what one outlier will do before I drag it."
Tips for the workforce loop
Verbs in your prompt shape what the Agent ships
"Mess with," "watch what happens," "feel why" — that language is what gets a manipulable canvas. The Tutor reads interaction verbs as a signal to dispatch a build-and-drag Agent; a plain "explain" tends to dispatch a passive demo Agent instead.
Tell the Tutor what to adjust and the Agent reships
If the first ship is close but missing something — a second variable, a simpler version, a toggle — say so. The Tutor reissues the brief and the Agent rebuilds. The back-and-forth is how you land the version that fits what you're trying to feel.
What to do with the visual next
Hover for options: copy as image drops it into your notes, Save as Artifact keeps the canvas interactive for later. Or ask the Tutor to write up what you now understand — the visual surfaced one gap, putting it in words often surfaces the next.