正在打开每日关注
正在打开每日关注
通过多智能体协作自动化日常工作,探索 Agent 工程的落地方法。
God damn. I cannot recommend more strongly that everyone, EVERYONE, whether you’re an Eng, pm, or even investor/finance bro, watch this. At LEAST the first half. @poteto was on the react compiler team at meta and an EM at Netflix. What struck me most is the CLEAR skepticism she once had about agents, how she talks through all the problems IN DETAIL, but how she systematized solving those problems and sort of climbed up the hill of agent incompetence despair.
SpaceXAI engineer (ex-Cursor): "right now I'm running 10-20 GrokBot agents that automate 90% of my routine i have a Chief of Staff agent. He knows about all my other bots and manages everything" in a 50-minutes podcast, a SpaceXAI engineer showed how to build a team of agents that will work for you 24/7 worth more than a $500 course on agentic engineering watch today, then read how to build a Grok agents team from scratch in the article below
是什么 这是一条对 Agent 工程访谈内容的强烈推荐。核心主题不是单个模型或产品,而是如何把多个 Agent 组织成可持续运行的工作系统:由不同 Bot 承担日常任务,再由一个类似 Chief of Staff 的 Agent 统一了解、协调和管理其他 Bot。转推内容还给出了“10–20 个 Agent 自动化 90% 例行工作”的个人案例。
核心设计 其价值在于把 Agent 从一次性聊天工具提升为可分工、可调度、可管理的协作网络。原作者特别强调,嘉宾曾经对 Agent 持怀疑态度,并详细讨论其能力不足,随后通过系统化方法逐步解决问题。这暗示落地关键不只是提示词,而是任务拆解、角色分工、协调机制、持续校正与人机边界。
关键洞察 Agent 的成熟路径可能是从单点自动化开始,再逐步形成团队化编排;管理层、产品和工程人员都需要理解这种变化,因为它会重塑个人生产方式与组织协作方式。
边界与风险 “自动化 90%”和“24/7 工作”属于转述中的个人经验,缺少任务范围、质量指标、成本和失败率,不能直接视为普遍结论。文本也没有提供具体架构、评测方法或安全控制细节,因此更适合作为学习入口,而非可直接复制的实施方案。