CrewScaler 16-protocol AI-tutor coaching library

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 causal), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The framework specifies the AI tutor's pedagogy in advance as a library of 16 named protocols drawn from the retrieval-practice, productive-failure, and affect literatures, rather than a single general-purpose prompt. Protocols include direct explanation, Socratic questioning, worked examples, hint escalation, spaced retrieval, productive failure, and "affective support that prioritizes boredom over frustration". A selection layer picks the active protocol each turn by fixed priority: integrity risk first, then affective signals, then group context, then learner intent. Six cross-cutting principles constrain every protocol.

Design Implications

Context

Requirements

• A selection layer that prioritizes integrity risk, affective signals, group context, and learner intent
• Grounded RAG and academic-integrity guardrails as stated quality controls

Constraints

• The article cites evidence that current LLMs labeled incorrect learner actions no better than chance in a 223-domain testbed, motivating explicit protocol design

Target Learners

• Adult professionals in frontier technical upskilling programs

Learning Goals

• Reducing time-to-competency through protocolized one-to-one tutoring

Claims

• Prompt specificity changes AI tutor pedagogy: a no-direct-answer guardrail alone produces uniform questioning, while a detailed guided protocol produces varied, impasse-sensitive responses [+W]

Related Designs

• Retrieval-augmented Conversational AI Tutor with modular GROW-aligned coaching routines
• CrewScaler: an end-to-end five-stage AI-accelerated upskilling pipeline

Examples

-

Key Sources

• Tam Nguyen, Hung Nguyen, Robert Ogburn. (2026). AI-accelerated End-to-End Framework for Rapid Professional Upskilling. https://arxiv.org/abs/2607.14044

Page metadata

Field	Value
type	design
id	crewscaler-16-protocol-ai-tutor-library
title	CrewScaler 16-protocol AI-tutor coaching library
description	The framework specifies the AI tutor's pedagogy in advance as a library of 16 named protocols drawn from the retrieval-practice, productive-failure, and affect literatures, rather than a single general-purpose prompt.
status	review
generated	by process:settle-candidates on 2026-10-10
trust	unverified — no one has confirmed this page's content
sources	1 — tam-nguyen-2026

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