Phase 19: Capstone Projects

卡普斯通17 个人人工智能导师 (适应性,多模特,具有内存)

米戈 (汗学院),杜林戈马克斯,教育谷歌学习LM/双胞胎,奇威letQ-聊天,和合成导师都在2026年提供了适应性多模式导师. 常见的形式是苏格拉底政策 (永远不要只是抛弃答案),学习者模型每次交互后都会更新 (贝叶斯知识追踪风格),语音+文字+照片数学输入,课程图图检索,间隔重复计划和适合年龄的安全过器. 终点是向专科的导师 (K-12代数或Python引入),与10名学习者进行两周的有效性研究,并通过内容安全审计.

Type: Capstone

Languages: Python (backend, learner model), TypeScript (web app), SQL (curriculum graph via Postgres + Neo4j)

Prerequisites: Phase 5 (NLP), Phase 6 (speech), Phase 11 (LLM engineering), Phase 12 (multimodal), Phase 14 (agents), Phase 17 (infrastructure), Phase 18 (safety)

Phases exercised:五·六·十一·十二·十四·十七·十八

Time: 30 hours

问题

适应教学曾经是科技研究领域的位. 到2026年,它将成为消费品. 美国大多数学校都部署了Khanmigo. 杜灵戈马克斯打了数亿的MAU. 谷歌的学习LM/双子座教育功能在谷歌课堂教学. 问卷Q-Chat坐在卡片旁边. 导师与导师为好奇的孩子们的传播. 共同的元素:多模式输入 (类型,讲话,摄影方程),苏格拉底教学 (先问,后解释),每次互动后更新的学习模型,以及严格的适合年龄的安全.

测量是实际有效性研究:测试前和测试后两周的成绩,包括10名学习者.语音循环必须感觉自然 (Capstone 03子堆).记忆必须尊重隐私.安全过器必须通过Coppa意识的红队K-12.

概念

它们有四个组成部分.Tutor policy学生问答时,政策提出一个主要问题;当他们正确地理解时,它转向下一个概念;当他们陷入困境时,它提供了一个架的提示. Learner model是贝叶斯知识追踪 (或简单的变体) 每次交互后更新每一个课程节点的掌握概率. Curriculum graph对于一个概念的定义,该概念的定义是:Memory是一个节目式+语义存储 (代理记忆式) 包含过去的互动,错误和偏好.

通过Dots.ocr或PaliGemma进行数学问题的照片输入.通过Cartesia Sonic-2进行语音输出.安全使用Llama Guard 4加上年龄适当的过器 (阻止成人内容,暴力,自伤) 和COPPA意识的记忆保留政策.

效率研究是可交付的. 10名学习者,试验前和试验后,两周.报告学习增长的特拉和信心间隔.与非适应的基线 (不需要导师政策的线性内容) 进行比较.

建筑

learner device
  |
  +-- text         -> web app
  +-- voice        -> LiveKit Agents (ASR + TTS)
  +-- photo math   -> dots.ocr / PaliGemma 2
       |
       v
  tutor policy (LangGraph)
       - Socratic decision head
       - next-concept chooser (curriculum graph walk)
       - hint scaffolder
       - mastery update
       |
       v
  learner model (BKT / item-response theory)
       - per-concept mastery probability
       - spaced-repetition scheduler (SM-2 or FSRS)
       |
       v
  memory (agentmemory-style)
       - episodic: every interaction
       - semantic: learned mistakes, preferences
       - retention policy: COPPA / GDPR aware
       |
       v
  curriculum graph (Neo4j)
       - prerequisite edges
       - OER content attached
       |
       v
  safety:
    Llama Guard 4 + age-appropriate filter
    memory access guarded by learner ID scope

堆

选择主题:K-12代数或Python介绍 (选择一个深度)
导师政策:LangGraph与Claude Sonnet 4.7 (即时缓存)
学习者模型:贝耶斯知识追踪 (经典) 或FSRS用于间隔
课程图:概念的新4j+先决条件边缘+OER内容
记忆:代理记忆式持久向量+剧情+语义存储
声音:LiveKit代理 1.0 +卡特西亚索尼克-2 (重复使用的底石03子堆)
图形数学:dots.ocr或PaliGemma 2用于识别方程
安全:Llama Guard 4+适合年龄的定制过器
率:率问题生成,测试前/后使用,有效性研究工具

建立它

Curriculum graph.构建一个由50-150个概念节点组成的Neo4j (例如,从"数线"到"方程公式"的K-12代数) 具有先决条件边缘. 附加每个节点的OER内容 (Open Textbook, OpenStax).

Learner model.开始使用先例来追踪贝耶斯知识:猜测,滑动,学习速度. 每次交互后更新每个概念的掌握. 持续每个学习者.

Tutor policy.带节点的长图: read_signal(学习者的答案是否正确/部分/固?),select_concept(步行课程图选择最优先概念),scaffold现在,我们需要一个人来帮助我们.update_mastery现在,我们要去.

Memory.任何交互都会写到一个剧情商店. 错误和偏好促进语义记忆. COPPA 意识到保留政策:自动删除1年后,可以访问父母.

Voice path.通过Whisper-v3-turbo.TTS通过Cartesia Sonic-2. 支持轮 (重复使用 capstone 03 机械).

Photo-math path.运行dots.ocr或PaliGemma 2以识别方程; 输送给导师作为结构化输入.

Safety.每个模型输出都通过了Llama Guard 4 + 适合年龄的过器 (阻止自伤,成人内容,暴力).学习者ID的内存访问范围;删除的父母访问表面.

Efficacy study.10名学习者,预测 (30个标准问题基础),两周的导师互动 (3个会议/周),测试后.

Weekly progress reports.根据学习者,自动生成已探讨的主题,掌握轨迹和建议的下一步的PDF总结.

用它

learner: "I don't understand why 3x + 6 = 12 means x = 2"
[signal]   stuck
[concept]  'isolating variables' (prerequisite: addition-subtraction-equality)
[scaffold] "what number would you subtract from both sides to start?"
learner: "6"
[signal]   correct
[mastery]  addition-subtraction-equality: 0.62 -> 0.77
[concept]  continue 'isolating variables'
[scaffold] "great. now what is 3x / 3 equal to?"

运送它

outputs/skill-ai-tutor.md具有多模式输入,学习模型,记忆,安全性和测量效率的专科特定适应教师.

Weight	Criterion	How it is measured
25	Learning gain delta	Pre/post-test delta in a 10-learner two-week study
20	Socratic fidelity	Rubric score on transcript samples
20	Multimodal UX	Voice + photo + text coherence end to end
20	Safety + privacy posture	Llama Guard 4 pass rate + COPPA-aware retention
15	Curriculum breadth and graph quality	Concept coverage + prerequisite graph consistency
100

运动

运行有效性研究,无论是没有适应性学习者模型 (随机概念顺序).报告 delta. 预计适应性赢得,但大小是有趣的数字.

添加多模特探测器:相同的概念问题作为文字,声音和照片. 测量学习者是否更快地与他们喜欢的模式相结合.

建立一个主题仪表板:练习的主题,掌握轨迹,即将推出的概念,安全事件 (任何防护车道.

增加语言交换模式:导师接受西班牙语输入,并以西班牙语教学.

强调记忆隐私:验证学习者A即使通过语音录像重新摄入攻击,也无法看到学习者B的数据.记录访问尝试和警报.

关键词

Term	What people say	What it actually means
Socratic policy	"Ask, do not dump"	Tutor asks a leading question rather than giving the answer
Bayesian knowledge tracing	"BKT"	Classic learner-model equations for mastery probability per concept
FSRS	"Free Spaced Repetition Scheduler"	2024 spaced-repetition scheduler, better than SM-2
Curriculum graph	"Concept DAG"	Neo4j of concepts with prerequisite edges
Episodic memory	"Per-interaction log"	Every interaction stored for later retrieval
Semantic memory	"Learned pattern store"	Compacted mistakes and preferences promoted from episodic
COPPA	"Kids privacy law"	US law restricting data collection from children under 13

进一步阅读

Khanmigo (Khan Academy)参考消费者K-12教师
Duolingo Max参考语言学习教师
Google LearnLM / Gemini for Education托管的参考模型
Quizlet Q-Chat替代参考
Synthesis Tutor创业参考
FSRS algorithm间隔重复时间表
Bayesian Knowledge Tracing学习者模型经典
LiveKit Agents语音堆

This free lesson is part of the AI Engineering from Scratch curriculum. Read the full explanation, run the lesson code, and verify the result in the interactive reader or from the repository source.

Browse the complete course catalog or open this lesson on GitHub.