Teacher Interaction Layer: a web-based configuration surface for pedagogical intent, persona, and role¶

Design · All designs
Evidence · no claims cited

Description¶

The Teacher Interaction Layer is a lightweight web-based configuration page through which educators shape the virtual tutor's behavior; all settings are injected directly into the LLM system prompt. It exposes personality presets (e.g., Neutral, Friendly, Patient, Verbose, Concise, plus a Custom option), thumbnail-based avatar appearance selection, a knowledge base selector, and teacher type. "The interface specifies the type of teacher through mutually exclusive radio-button options (e.g., professor (replicating lecture content), tutor (providing repeated explanations and scaffolding), and exam trainer", with a save action persisting settings as reproducible configuration artifacts.

Design Implications¶

Context¶

Requirements¶

Settings must be persisted as configuration artifacts (templates, rule sets, corpus descriptors) interpreted by the server at runtime for traceability

Constraints¶

Personality presets are deliberately constrained; styles beyond them require additional engineering inputs such as tailored prompt templates or policy rules

Target Learners¶

students in classrooms configured by their teachers

Learning Goals¶

consistent, pedagogically framed tutoring behavior matching classroom norms and assessment contexts

Claims¶

TASD: teacher-authored prompt layers as orchestration levers for configuring student-AI dialogue

Examples¶

-

Key Sources¶

Lorenzo Stacchio, Michele Giordano, Daniele Berardini, Primo Zingaretti, Emanuele Frontoni. (2026). ELEVATE: Designing Human-Centered GenAI Virtual Tutors for Scalable and Inclusive Education. arXiv preprint. https://arxiv.org/abs/2606.30662

Page metadata

Field	Value
`type`	`design`
`id`	`elevate-teacher-interaction-layer`
`title`	`Teacher Interaction Layer: a web-based configuration surface for pedagogical intent, persona, and role`
`description`	`The Teacher Interaction Layer is a lightweight web-based configuration page through which educators shape the virtual tutor's behavior; all settings are injected directly into the LLM system prompt.`
`status`	`review`
`generated`	by `process:settle-candidates` on `2026-10-10`
`trust`	unverified — no one has confirmed this page's content
`sources`	1 — `lorenzo-stacchio-2026`

Frontmatter is the page's machine-readable record — see the Schema & Guide.