←News detail

East MoneyUpdated Nov 1 09:24

DeepSeek Harness team publishes exploration of Super-agent architecture design

English text is being prepared.

AI summary

The DeepSeek Harness team published a technical article titled "Super-agent Design Exploration: Toward a More Autonomous Agent System" on its official WeChat account, proposing a dual-layer architecture, event-driven design, asynchronous multi-channel interaction, separation of storage and compute, and an external Watchdog that takes over session management and context synchronization, shifting the multitasking ceiling from human attention to the system layer. The team says existing Agents operate as single Sessions that are independent of one another, and a person can typically follow only 3 to 5 active Sessions at the same time. Based on this approach and DeepSeek Harness, an internal Super-agent instance has been running continuously, participating in issue reproduction and fixes, code review responses, and iteration on Harness itself; the team emphasizes that this is an engineering-practice-oriented discussion of architectural ideas, not an introduction to existing product features.

Why read this

AI score 60

The DeepSeek Harness team has publicly disclosed the dual-layer architecture and event-driven design of Super-agent, and stated that internal instances have already been put into daily development workflows.

Topics

Tags

Body

文章主体

10月31日晚,DeepSeek Harness团队在官方公众号发表技术文章《Super-agent设计探索:迈向更自主的Agent系统》。

DeepSeek表示,模型能力提升催生了新的应用形态,个人助手(Personal Agent)开始集中涌现。它们直接接入日常的消息、日程、邮件与工作平台,持续观察信息流并协助处理事务。这种持续在线、事件驱动的工作模式并不局限于个人场景,个人助手只是其中一类应用;在软件开发等商业场景中,它同样很有价值。

然而,现有主流 Agent设计还难以在多任务和团队协作的规模上支撑这一模式,多任务上限取决于人的注意力。

Agent的工作模式仍然以单一Session为基本单位,不同Session之间彼此独立。当用户并行推进多项工作时,必须手动切换并记住各个Session的上下文,而据团队观察,一个人通常只能同时跟进3到5个活跃Session。在这种设计下,单任务的上限取决于模型能力,多任务的上限取决于人的注意力。

DeepSeek团队通过双层架构、事件驱动、异步多信道交互、存算分离与外部Watchdog,系统在底层接管了繁琐的会话管理与上下文同步,使人类得以从具体的执行细节进一步上移到更高层的决策与验收。

DeepSeek把这种Agent形态及其系统架构称为Super-agent。

文章指出,整套设计并没有提出全新的理论,而是将计算机系统中经典的工程思想应用到了Agent系统的构建中。具体的实现方式会随技术发展而演进,但DeepSeek认为背后的系统设计思路有更长期的参考价值。

目前,DeepSeek基于这套思路和DeepSeek Harness构建的内部Super-agent实例已经在持续运行,日常参与问题复现与修复、代码审查响应与DeepSeek Harness自身的迭代。

值得注意的是,在本篇推文的开始,DeepSeek Harness团队也提示,这是一篇面向工程实践的架构思路探讨,并非对DeepSeek Harness现有产品功能的介绍。

(文章来源:每日经济新闻) 正文中部 内嵌广告

Source

East Money