当前位置：首页 > news >正文

面向教育 Agent 的 Harness 学习进度持久化

news 2026/8/22 8:40:40

面向教育 Agent 的 Harness 学习进度持久化落地全指南

作者：老周 | 资深AI教育产品技术负责人 | 深耕教育Agent落地2年
本文字数：14800+字 | 建议收藏后分段阅读，配套代码可直接用于生产环境

引言

痛点引入

我相信所有做过教育Agent落地的开发者都踩过「学习进度持久化」的坑：

测试用户反馈用手机刷了2天初二数学题，换电脑登录之后进度全丢，排查发现是之前用内存存会话状态，服务重启数据全部清零，被老板骂了整整一下午；
多端同时学习时进度冲突，用户在平板上做了一套测评掌握度升到0.7，手机上做了一道题进度回退到0.5，用户投诉说自己的学习成果被「吃了」；
迭代知识点图谱时存量进度不兼容，我们把初中数学的「一元一次方程」拆成了3个子知识点，老用户的进度全部失效，不得不连夜写迁移脚本，还赔了用户1个月的会员；
合规检查不过关，教育数据属于敏感个人信息，我们之前把用户学习记录存在境外的云服务上，被监管警告，差点停服整改。

这些问题本质上不是业务逻辑的问题，而是教育Agent的持久化基础设施没有适配场景特性：普通的CRUD存储解决不了多端冲突、版本兼容、合规审计、高可用扩展这些教育场景的刚性需求。

解决方案概述

我们团队花了2个月时间，基于Harness软件交付平台重构了整个教育Agent的持久化层，上线3个月以来，进度同步成功率达到99.998%，没有出现过一次进度丢失、冲突的问题，开发成本比自研降低了70%，运维成本降低了90%，还一次性通过了等保2.0三级认证。

这套方案的核心优势：

开箱即用的基础设施：不需要自己从零实现缓存、多存储适配、灰度发布、合规校验等能力，Harness已经全部封装好了；
冲突-free的多端同步：内置CRDT算法，支持无锁的多端进度合并，不会丢失任何学习行为数据；
版本兼容保证：基于Harness Schema注册中心，知识点图谱迭代时自动兼容存量进度数据，不需要手动写迁移脚本；
合规强制生效：基于Harness Policy as Code，所有数据操作都符合《个人信息保护法》《教育数据安全规范》的要求，不需要担心合规风险；
弹性扩缩容：用户量从10万涨到100万不需要改一行代码，Harness自动调整存储资源配额。

最终效果展示

我们的K12数学教育Agent上线这套方案之后的数据表现：

多端进度同步延迟<1s；
进度数据准确率100%，没有出现过一次冲突或丢失；
知识点迭代上线时间从3天缩短到2小时，不需要做数据迁移；
合规审计100%通过，不需要额外投入合规研发资源。

准备工作

环境/工具要求

工具/环境	版本要求	用途说明
Harness Platform	SaaS版 / 本地部署版≥v0.25	提供持久化抽象、策略管理、版本管理、灰度发布能力
Python	≥3.10	教育Agent业务逻辑开发
LangChain	≥0.1.0	Agent编排框架，配套回调实现自动持久化
PostgreSQL	≥14	结构化学习进度数据存储
Redis	≥7.0	热学习进度缓存，降低延迟
Pinecone / Milvus	≥2.0	用户交互历史向量存储，支持学习行为溯源
阿里云OSS / 腾讯云COS	不限	冷数据归档、用户作业等大文件存储

前置知识要求

阅读本文需要你具备以下基础知识：

教育Agent的基本架构，了解LangChain的基本使用；
基本的数据库操作知识，了解关系型数据库、缓存、向量数据库的适用场景；
基本的分布式系统概念，了解最终一致性、冲突解决的基本原理；
对Harness平台有基本了解，不需要精通，跟着本文步骤操作即可。

核心概念与问题背景

核心概念定义

1. 教育Agent

教育Agent是基于大语言模型构建的、能够为用户提供个性化学习服务的智能体，核心能力包括知识点讲解、错题分析、自适应刷题、测评诊断、学习路径规划等，核心特点是和用户的交互是会话式的、连续的，需要基于用户的历史学习状态提供个性化服务。

2. 学习进度的核心要素

教育场景的学习进度不是一个简单的「学到第几章」的字段，而是包含多维度数据的结构化对象，核心要素如下：

要素	类型	说明
主体标识	字符串	用户唯一ID、会话ID、客户端标识
知识点关联	字符串	对应的知识点ID、知识点版本、所属学科和年级
掌握状态	浮点数	知识点掌握度（0-1）、累计学习时长、做题数、正确率
历史链路	结构化对象	交互历史、错题记录、测评得分、停留时长
版本信息	整数	数据模型版本、知识点图谱版本

3. Harness平台核心组件

本文用到的Harness核心组件如下：

Harness Persistence Layer：统一的持久化抽象层，支持对接各类存储源，屏蔽底层存储的复杂度；
Harness Schema Registry：数据模型注册中心，自动校验Schema变更的兼容性，避免Breaking Change；
Harness Policy as Code：策略即代码引擎，可自定义合规规则，所有数据操作都要经过规则校验才能执行；
Harness Feature Flags：特性开关，支持灰度发布新的持久化逻辑，出问题可一键回滚；
Harness Event Bus：事件总线，支持进度更新的实时推送，实现多端同步。

4. 教育场景持久化的核心要求

和普通的业务数据持久化不同，教育场景的持久化有以下刚性要求：

不可丢失性：任何学习行为数据都不能丢失，否则会影响用户的学习体验和信任；
多端一致性：三端（Web/APP/小程序）的进度必须保持一致，同步延迟<1s；
可溯源性：支持查询用户任意时间段的学习历史，用于学习分析和错题溯源；
版本兼容性：知识点图谱迭代时，存量进度数据必须自动兼容，不能失效；
合规性：符合《个人信息保护法》《教育数据安全规范》的要求，数据必须存在境内，用户有权删除所有个人数据；
高可用性：持久化服务的可用性必须≥99.99%，不能因为存储故障影响Agent服务。

行业发展历史与现存痛点

我们整理了近15年教育产品持久化技术的演变历史，如下表所示：

时间	教育产品形态	持久化技术栈	核心痛点	主流解决方案
2010年及以前	单机教育软件、光盘课程	本地文件、Access数据库	换设备进度丢失，数据容易损坏	手动导出导入进度文件
2015-2019年	在线教育平台、直播课	MySQL、云服务器存储	同步延迟高，扩展困难，运维成本高	读写分离、分库分表
2020-2022年	AI助教、自适应学习系统	MySQL+Redis+MongoDB	多端冲突，合规性差，迭代兼容性差	分布式事务、最后写入获胜策略
2023年至今	生成式教育Agent	自研持久化层+多模存储	学习行为数据溯源难，跨场景进度复用难	手动实现冲突解决、版本管理
2025年（预测）	全域教育Agent网络	联邦学习+分布式身份+跨链存储	数据孤岛，用户对数据没有控制权	学习进度跨平台互通、用户自主掌控数据

当前教育Agent持久化的普遍痛点：

临时存储导致数据丢失：很多团队为了快速上线，用内存或本地文件存会话状态，服务重启、设备切换就会丢失进度；
冲突解决策略不合理：大多采用「最后写入获胜（LWW）」的策略，会丢失更早的学习行为数据，比如用户在两个端同时学习，只会保留最后一个端的进度；
版本兼容能力缺失：知识点图谱迭代时，存量进度数据不兼容，需要手动写迁移脚本，容易出问题；
合规能力不足：教育数据属于敏感信息，很多团队没有完善的合规机制，容易违反监管要求；
运维成本高：需要自己维护多套存储、缓存、同步服务，用户量上涨时还要手动扩缩容，运维成本极高。

整体架构设计

系统架构总览

我们的整体架构分为4层，如下图所示：

渲染错误:Mermaid 渲染失败: Parsing failed: Lexer error on line 2, column 15: unexpected character: ->(<- at offset: 32, skipped 11 characters. Lexer error on line 3, column 20: unexpected character: ->(<- at offset: 63, skipped 1 characters. Lexer error on line 3, column 24: unexpected character: ->端<- at offset: 67, skipped 2 characters. Lexer error on line 4, column 20: unexpected character: ->(<- at offset: 97, skipped 1 characters. Lexer error on line 4, column 24: unexpected character: ->端<- at offset: 101, skipped 2 characters. Lexer error on line 5, column 24: unexpected character: ->(<- at offset: 135, skipped 6 characters. Lexer error on line 7, column 16: unexpected character: ->(<- at offset: 170, skipped 3 characters. Lexer error on line 7, column 24: unexpected character: ->层<- at offset: 178, skipped 3 characters. Lexer error on line 7, column 29: unexpected character: ->教<- at offset: 183, skipped 2 characters. Lexer error on line 7, column 36: unexpected character: ->]<- at offset: 190, skipped 1 characters. Lexer error on line 8, column 24: unexpected character: ->(<- at offset: 215, skipped 1 characters. Lexer error on line 8, column 28: unexpected character: ->网<- at offset: 219, skipped 3 characters. Lexer error on line 9, column 30: unexpected character: ->(<- at offset: 261, skipped 1 characters. Lexer error on line 9, column 40: unexpected character: ->编<- at offset: 271, skipped 3 characters. Lexer error on line 10, column 20: unexpected character: ->(<- at offset: 303, skipped 7 characters. Lexer error on line 11, column 25: unexpected character: ->(<- at offset: 344, skipped 9 characters. Lexer error on line 13, column 18: unexpected character: ->(<- at offset: 385, skipped 1 characters. Lexer error on line 13, column 26: unexpected character: ->中<- at offset: 393, skipped 5 characters. Lexer error on line 13, column 38: unexpected character: ->平<- at offset: 405, skipped 4 characters. Lexer error on line 14, column 24: unexpected character: ->(<- at offset: 433, skipped 8 characters. Lexer error on line 15, column 23: unexpected character: ->(<- at offset: 475, skipped 1 characters. Lexer error on line 15, column 30: unexpected character: ->注<- at offset: 482, skipped 5 characters. Lexer error on line 16, column 23: unexpected character: ->(<- at offset: 521, skipped 9 characters. Lexer error on line 17, column 19: unexpected character: ->(<- at offset: 560, skipped 1 characters. Lexer error on line 17, column 32: unexpected character: ->)<- at offset: 573, skipped 1 characters. Lexer error on line 18, column 25: unexpected character: ->(<- at offset: 610, skipped 6 characters. Lexer error on line 19, column 19: unexpected character: ->(<- at offset: 646, skipped 6 characters. Lexer error on line 21, column 18: unexpected character: ->(<- at offset: 686, skipped 13 characters. Lexer error on line 22, column 22: unexpected character: ->(<- at offset: 721, skipped 1 characters. Lexer error on line 22, column 29: unexpected character: ->热<- at offset: 728, skipped 6 characters. Lexer error on line 23, column 19: unexpected character: ->(<- at offset: 764, skipped 1 characters. Lexer error on line 23, column 31: unexpected character: ->结<- at offset: 776, skipped 6 characters. Lexer error on line 24, column 25: unexpected character: ->(<- at offset: 818, skipped 1 characters. Lexer error on line 24, column 35: unexpected character: ->向<- at offset: 828, skipped 5 characters. Lexer error on line 25, column 20: unexpected character: ->(<- at offset: 864, skipped 1 characters. Lexer error on line 25, column 25: unexpected character: ->冷<- at offset: 869, skipped 6 characters. Parse error on line 3, column 21: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Web' Parse error on line 3, column 27: Expecting token of type ':' but found `in`. Parse error on line 4, column 21: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'APP' Parse error on line 4, column 27: Expecting token of type ':' but found `in`. Parse error on line 7, column 19: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Agent' Parse error on line 7, column 27: Expecting token of type ':' but found `AI`. Parse error on line 7, column 31: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Agent' Parse error on line 7, column 37: Expecting token of type ':' but found ` `. Parse error on line 8, column 25: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'API' Parse error on line 8, column 32: Expecting token of type ':' but found `in`. Parse error on line 9, column 31: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'L' Parse error on line 9, column 44: Expecting token of type ':' but found `in`. Parse error on line 13, column 19: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Harness' Parse error on line 13, column 31: Expecting token of type ':' but found `Harness`. Parse error on line 15, column 24: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Schema' Parse error on line 15, column 36: Expecting token of type ':' but found `in`. Parse error on line 17, column 20: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Feature' Parse error on line 17, column 28: Expecting token of type ':' but found `Flag`. Parse error on line 17, column 34: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'in' Parse error on line 17, column 44: Expecting token of type ':' but found ` `. Parse error on line 22, column 23: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'R' Parse error on line 22, column 36: Expecting token of type ':' but found `in`. Parse error on line 23, column 20: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'PostgreSQL' Parse error on line 23, column 38: Expecting token of type ':' but found `in`. Parse error on line 24, column 26: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'Pinecone' Parse error on line 24, column 41: Expecting token of type ':' but found `in`. Parse error on line 25, column 21: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: 'OSS' Parse error on line 25, column 32: Expecting token of type ':' but found `in`. Parse error on line 27, column 9: Expecting token of type 'ARROW_DIRECTION' but found `80`. Parse error on line 27, column 12: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: '--' Parse error on line 27, column 23: Expecting token of type ':' but found ` `. Parse error on line 28, column 9: Expecting token of type 'ARROW_DIRECTION' but found `80`. Parse error on line 28, column 12: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: '--' Parse error on line 28, column 23: Expecting token of type ':' but found ` `. Parse error on line 29, column 13: Expecting token of type 'ARROW_DIRECTION' but found `80`. Parse error on line 29, column 16: Expecting: one of these possible Token sequences: 1. [NEWLINE] 2. [EOF] but found: '--' Parse error on line 29, column 27: Expecting token of type ':' but found ` `. Parse error on line 30, column 13: Expecting token of type ':' but found `--`. Parse error on line 30, column 17: Expecting token of type 'ARROW_DIRECTION' but found `orchestration`. Parse error on line 31, column 19: Expecting token of type ':' but found `--`. Parse error on line 31, column 23: Expecting token of type 'ARROW_DIRECTION' but found `llm`. Parse error on line 32, column 19: Expecting token of type ':' but found `--`. Parse error on line 32, column 23: Expecting token of type 'ARROW_DIRECTION' but found `callback`. Parse error on line 33, column 14: Expecting token of type ':' but found `--`. Parse error on line 33, column 18: Expecting token of type 'ARROW_DIRECTION' but found `adapter`. Parse error on line 34, column 13: Expecting token of type ':' but found `--`. Parse error on line 34, column 17: Expecting token of type 'ARROW_DIRECTION' but found `schema`. Parse error on line 35, column 13: Expecting token of type ':' but found `--`. Parse error on line 35, column 17: Expecting token of type 'ARROW_DIRECTION' but found `policy`. Parse error on line 36, column 13: Expecting token of type ':' but found `--`. Parse error on line 36, column 17: Expecting token of type 'ARROW_DIRECTION' but found `ff`. Parse error on line 37, column 13: Expecting token of type ':' but found `--`. Parse error on line 37, column 17: Expecting token of type 'ARROW_DIRECTION' but found `redis`. Parse error on line 38, column 13: Expecting token of type ':' but found `--`. Parse error on line 38, column 17: Expecting token of type 'ARROW_DIRECTION' but found `pg`. Parse error on line 39, column 13: Expecting token of type ':' but found `--`. Parse error on line 39, column 17: Expecting token of type 'ARROW_DIRECTION' but found `pinecone`. Parse error on line 40, column 13: Expecting token of type ':' but found `--`. Parse error on line 40, column 17: Expecting token of type 'ARROW_DIRECTION' but found `oss`. Parse error on line 41, column 14: Expecting token of type ':' but found `--`. Parse error on line 41, column 18: Expecting token of type 'ARROW_DIRECTION' but found `web`. Parse error on line 42, column 14: Expecting token of type ':' but found `--`. Parse error on line 42, column 18: Expecting token of type 'ARROW_DIRECTION' but found `app`. Parse error on line 43, column 14: Expecting token of type ':' but found `--`. Parse error on line 43, column 18: Expecting token of type 'ARROW_DIRECTION' but found `miniapp`. Parse error on line 44, column 8: Expecting token of type ':' but found `--`. Parse error on line 44, column 12: Expecting token of type 'ARROW_DIRECTION' but found `adapter`.

各层的职责：