A16z-Backed Cosign Opens Reputation Graph to Personal AI Agents

a16z投资的Cosign向个人AI Agent开放声誉图谱 传统招聘网络之所以失效，是因为公开简历缺乏信任信号，促使Cosign构建了一个可供Agent查询的图谱，让个人AI能够基于同行背书促成直接匹配。 David Booth与Dhruv Gupta与主持人Sophia Dew一同介绍了Cosign及首届Cosign Awards。该平台旨在将专业人才发掘方式从冷冰冰的简历转向高信任度的同行背书。讨论剖析了Cosign的分层架构——涵盖基础档案、精选名单和公开背书——并强调了其成为创业生态系统中面向Agent的原生事实来源的抱负。展望未来，团队设想了一种Agent对Agent的社交网络范式，个人AI助手可以直接与招聘经理或公司进行协商，仅在确认双方意图吻合时才促成真人接触。 a16z New Media团队介绍了Cosign，这是一个关系与声誉网络，旨在通过经同行验证的背书和Agent原生图谱搜索，取代充斥杂音的简历招聘模式。 (0:00) Cosign的产品架构建立在三个层级之上：4万至5万份维基百科风格的基础档案、400多个精选名单，以及被称为“cosign”的高确信度同行背书。 (13:46) 为实现Agent之间的专业人脉发掘，Cosign正在集成模型上下文协议（MCP）端点、直接iMessage Agent以及对话式网络搜索工具。 (17:15) 首届Cosign Awards的提名和投票将持续至11月下旬，旨在表彰涵盖传统及互联网原生领域的从业者、后起之秀和产品。 (22:20)

Sophia Dew, David Booth, and Dhruv Gupta introduce Cosign, an initiative designed to replace noisy corporate recruiting with a peer-validated relationship graph and AI-mediated intent matching. Replacing Cold Applications with Peer Vouches (0:00) Traditional hiring funnels on platforms like LinkedIn and Handshake often subject applicants to low-signal resume filtering, while keeping skilled corporate workers bound by golden handcuffs due to limited visibility into early-stage startups. Gupta reflects on his own transition from finance cold applications to tech, which succeeded only after MTS CEO Brent explicitly vouched for his prior work at Polymarket. Cosign models professional discovery around high-conviction peer endorsements rather than static credentials. Booth defines a vouch as a personal stake where someone introduces a contact to an otherwise unreachable peer because the talent is genuinely worth their time. The platform aims to formalize these trust-based referrals so emerging operators can expand their serendipitous opportunities faster than standard job boards allow. Structuring Profiles, Lists, and the Cosign Awards (5:00) Cosign organizes its discovery system into three operational layers: an initial Wikipedia-style base layer indexing roughly 40,000 to 50,000 individual profiles, over 400 community-curated lists such as Packy McCormick's vertically integrated selections or historical Thiel Fellows rosters, and explicit, public peer vouches. The project seeks to make stealth startups and geographically distant talent visible, citing cases like a skilled 17-year-old developer in Sweden reaching out to MTS. To spotlight internal operators and talent density often overshadowed by celebrity founders, the team launched the Cosign Awards across people, companies, and products. The categories span standard functions like engineering and design alongside internet-native categories like "Public Intellectual of the Year" with 65 nominees, "Podcaster of the Year" with 76 nominees, and a "Rising Stars" group of 135 early founders and operators. Developing an Agent-Native Relationship Graph (17:15) Gupta contends that incumbent networks like LinkedIn keep their data siloed behind proprietary barriers and will not deploy open Model Context Protocol (MCP) tooling. After standard large language models like Google Gemini and ChatGPT returned generic recommendations when queried about local startup ecosystems, Cosign positioned itself as a structured relationship graph capable of processing taste, skills, and career intent. To make this data programmatically accessible, the platform provides an on-site search interface, an iMessage conversational agent, and MCP connectors for external AI workflows such as Muse, GrokBot, and Instinct. Booth notes that while web search is increasingly commoditized through tools like Exa, the defensible advantage in talent discovery lies in mapping private endorsement graphs rather than indexable public data. Enabling Autonomous Agent-to-Agent Alignment (20:00) The broader architectural vision involves autonomous agents negotiating directly on behalf of individuals and employers to establish a "coincidence of intent." While personal agents like Muse have expanded quickly to an estimated 10 million users, they currently remain limited to drafting outbound emails because they lack direct interaction protocols. Gupta notes that contemporary inbox design treats high-profile leaders like Sam Altman identically to teenagers, creating overwhelming inbound noise. Cosign's endpoint at cosign.co/agents is designed to let an applicant's agent interface directly with an enterprise or hiring manager's agent. The systems check for mutual goals and qualifications silently, passing the connection to humans only when direct alignment exists. Nominations for the platform's initial awards close in late November, with the team focusing on convening winners in person and refining onboarding workflows. Cosign's vision of autonomous, agent-mediated recruiting depends on widespread adoption of direct agent protocols and user willingness to publicly quantify peer reputations, leaving open whether external hiring managers will embrace decentralized graph signals over legacy credentialing.

Sophia Dew 与 a16z New Media 团队的 David Booth 和 Dhruv Gupta 探讨了其推出的职业社交项目 Cosign，提出利用基于信任的同行背书网络与 AI Agent 协议，打破传统招聘中的低信噪比与信息孤岛，实现自动化的“意图重合”匹配。 传统招聘的痛点与同行背书机制 (0:00) 传统职业求职渠道普遍依赖冷启动投递和简历关键字过滤。Dhruv Gupta 结合自身在宾夕法尼亚大学期间申请 Morgan Stanley 和 Jane Street 等机构的经历指出，在 Handshake 或 LinkedIn 等平台大量海投往往体验极差且效率低下。现行职业社交网络对所有人采用统一的收件箱机制，导致高价值人才被大量垃圾邮件淹没。与此同时，许多拥有出色专业能力的业务操盘手（operators）因“金手铐”或信息受限被困于大型企业，缺乏接触早期高使命感初创团队的途径。 Cosign 试图通过“同行背书”（vouching）构建高置信度的声誉信号，使专业人士的推荐权重超越单张简历。David Booth 指出，真正的核心信号在于某人是否愿意将候选人引荐给日程极其紧张且深度尊重的人士。团队推出首届 Cosign Awards，设置了包括软件工程、产品设计、AI 研究等核心职能，以及“年度公共知识分子”（65 位提名人）、“年度 X 平台发帖者”与“年度播客主”（76 位提名人）等互联网原生奖项，其中“新星”（Rising Stars）分类汇集了 135 位早期创始人与幕后骨干，旨在让公众关注从单纯的明星 CEO 转向实际驱动企业运转的关键人物。 Cosign 的三层产品架构与数据图谱 (13:46) 为了让隐形的高潜力初创公司与非核心创投中心（如瑞典等海外区域）的技术人才显性化，Cosign 将底层产品设计分为三个明确层次。第一层是维基百科式的底座层（Wikipedia-Style Base Layer），收录并搭建了约 40,000 至 50,000 名从业者、企业与投资人之间的关联档案；第二层是精选清单层（Curated Lists），汇聚了超过 400 个由社区生成的名单，例如 Packy McCormick 选出的六家垂直整合企业，或是 Thiel Fellows 与 Interact Fellows 历年入选者的追踪名录；第三层则是最高置信度的公开发帖背书（The Cosign），用于明确记录从业者愿意长期共事或早期发掘的人才。 David Booth 与 Dhruv Gupta 规划将 Cosign 打造成科技与初创生态系统的“真实信息源”（source of truth），让用户和组织能够检索基于图谱的复杂关系，例如查找特定区域的生物技术团队或追溯某位新兴人才的最早发掘者，从而系统化扩大个人遭遇机遇的“运气表面积”（surface area for luck）。 面向 AI Agent 的开放网络与 MCP 协议 (17:15) Dhruv Gupta 认为，传统职场平台（如 LinkedIn）将关系数据封存在专有壁垒内，永远不会提供支持广泛图谱检索的模型上下文协议（Model Context Protocol, 简称 MCP）。目前直接向通用大语言模型（如 Google Gemini 或 ChatGPT）询问特定城市的初创网络时，往往只能获得泛化且低价值的信息。随着 Exa 等专业搜索工具使通用网页搜索逐步商品化，真正的人才信息超额收益（alpha）将存在于私域的高置信度背书网络中。 为此，Cosign 将自身定位为 Agent 原生的网络基础设施，并部署了多项接入通道：包括网页内置的“Ask Cosign”查询工具、支持直接发短信检索图谱的 iMessage Agent，以及面向外部 AI 工作流（如 Muse、GrokBot 或 Instinct）的个人 MCP 接口。该架构旨在让个人 Agent 能够直接阅读结构化、富集后的关系层数据，根据用户的实际审美、技术专长和职业发展意图生成精准推荐。 Agent 间协商与意图重合工程 (20:00) 随着消费级个人智能体（例如用户量在数月内突破 1000 万的 Muse）快速普及，现有 Agent 缺乏有效的跨主体交互机制，大多停留在起草冷邮件等低效流程。在客户服务等领域，个人 Agent 与企业端 Agent 直接协商的协议已出现早期验证；Cosign 计划将该机制引入求职与业务协作网络中，通过 cosign.co/agents 开放端点支持意图匹配。 在该体系下，求职者的个人 Agent 可以在人类直接介入之前，直接对接招聘经理或潜在合作方的 Agent，双方对照技能、发展目标与职位需求进行前置对齐检查，自动化促成“意图重合”（coincidence of intent）。若双方目标一致，Agent 将促成进一步的人际连接；若不匹配，则进行静默过滤，从而消除低信噪比沟通对双方精力的消耗。 Cosign 当前的短期推进重点是完成首届 Cosign Awards 的社区评选，并计划在线下汇聚各类别获奖者以促进高信任度合作，同时团队正持续优化新用户入职引导与对话式网络情报功能。然而，该机制能否真正替代传统领英模式，仍取决于其声誉图谱能否在极高密度的硅谷圈层之外维持背书的真实性，以及跨 Agent 通信协议在更广泛招聘市场中的实际采纳率。

## Background 本期播客由 a16z New Media 团队的 David Booth、Dhruv Gupta 与 MTS 主持人 Sophia Dew 共同录制并发布于 RSS 播客平台（发布时间为 2026 年 10 月 31 日）。节目围绕新兴职业关系网络平台 Cosign 展开，重点介绍了其发起的 Cosign Awards，探讨如何摆脱传统简历与冷邮件投递，转而基于真实背书（Vouch）与 AI Agent 原生网络重塑职业连接。 ## TL;DR 节目深入探讨了传统职场招聘平台的低效痛点，剖析了机械海投与围墙花园生态如何掩盖真正的一线优秀操盘手。Cosign 通过“维基式主页、策展清单、深度背书”的三层数据架构，将隐性的熟人声誉沉淀为高置信度的人才图谱。展望未来，平台致力于构建 Agent 原生接口（如 MCP 与 iMessage），让个人 AI 智能体与企业智能体之间自主进行第一轮意图谈判，唯有在双方目标高度契合（Coincidence of Intent）时才促成人类连接。此外，团队介绍了首届 Cosign Awards 的评选机制，旨在挖掘被低估的幕后人才并促成高密度人才社群的线下聚合。 ## Atmosphere & style 对谈整体洋溢着浓厚的技术乐观主义与硅谷早期创业者的蓬勃活力，节奏轻快且充满默契。嘉宾们结合自身从金融海投到科技圈伯乐发掘的真实转折，既有对 LinkedIn 式“收件箱噪音”的犀利吐槽，也有对未来“Agent 替人类谈判找工作”的畅想与兴奋。整场对话既像是一场前沿产品原型的头脑风暴，又带着对职场中真诚人际纽带的温情敬意，听感真实、敏锐且极具前瞻性。 ## Bulletin - `00:00` Cosign 推出首届 Cosign Awards 奖项，旨在通过社区提名与专家评审，表彰幕后优秀操盘手（Operators）、高人才密度企业及标杆产品，投票与提名持续至 11 月底。 - `13:14` Cosign 平台构建了三层产品架构，目前已自动梳理并搭建了 4 万至 5 万个从业者维基式主页，并收录了约 400 份垂直领域的策展清单。 - `16:47` 团队正致力于打造 Agent 原生的职业网络，支持通过 MCP 协议、iMessage Agent 及内置对话接入外部 AI 模型，打破 LinkedIn 等传统平台的封闭壁垒。 - `20:00` 未来职业匹配将转向“Agent 对接 Agent”（Agent-to-Agent）协议机制，唯有在双方智能体确认存在真实的“意图契合”（Coincidence of Intent）时，才会促成人类面对面交流。 ## Key takeaways - `00:00` 传统简历投递与海投模式噪音大且效率低下，职业生涯中真正突破性的机遇高度依赖于行业信任背书（Vouch）。 - `00:00` 科技行业长久以来聚光灯多集中在创始人与 CEO 身上，Cosign 奖项旨在将业务一线发挥决定性作用的幕后操盘手推向台前。 - `05:00` Cosign 的终极愿景是成为初创生态的“真相之源”（Source of Truth），无论是人类还是 AI 智能体都能从中查询到高置信度的人才与人脉网络信息。 - `05:00` 平台的深层哲学是“扩大运气的接触面积”，通过结构化网络促成技能互补与抱负相通者的“意图重合”（Coincidence of Intent）。 - `10:00` 许多顶尖人才被困在薪酬丰厚但缺乏激情的“金手铐”传统岗位中，信任背书网络有助于帮他们完成职业冷启动，进入高密度人才圈子。 - `13:14` Cosign 产品包含三层核心逻辑：维基式结构化人才档案层、轻量级行业清单（Lists）策展层，以及代表深度声誉押注的 Cosign 背书层。 - `16:47` 传统职场社交巨头（如 LinkedIn）拥有极强的围墙花园属性，不会开放 MCP 让外部 AI 自由检索，导致高质量人才图谱数据在通用大模型中缺失。 - `16:47` 在人才市场获取超额收益（Alpha）依赖于挖掘未被公开标记的真实关系链与隐性信任数据，而非仅仅依赖公开的商品化数据。 - `20:00` 现有的职场社交平台收件箱机制缺乏有效过滤，导致领军人物与普通用户同样饱受低质冷邮件骚扰，社交信号严重失真。 - `20:00` 未来的求职与人脉拓展将由求职者个人 Agent 直接对接企业 Agent 展开第一轮意图谈判，达成契合后才引荐给人类决策者。 - `22:21` Cosign Awards 的核心价值不仅在于荣誉认证，更在于为各细分赛道的顶尖优胜者创造深度结识与面对面交流的高质量场域。 - `22:21` 团队目前正处于高强度快速迭代阶段，近期的产品开发重心在于重构入驻与上手引导（Onboarding）流程，以降低构建声誉网络的摩擦成本。 ## Detailed walkthrough ### `00:00` 打造基于信任背书的人脉网络与 Cosign 奖项的诞生 播客开篇描绘了一个前沿的求职设想：未来的求职不再依赖枯燥的简历投递，而是求职者个人的 AI 智能体（Agent）直接与企业方的智能体对接，自主开展第一轮意向匹配，评估彼此的目标和契合度，唯有在双方意图高度一致时才将机会推送到招聘经理面前。职业生涯中那些真正重大的机遇，往往取决于是否有人愿意为你公开背书（Vouch）。Cosign 的核心目标正是将这种强信任信号沉淀为一张全新的职业关系网络。MTS 主持人 Sophia 邀请到来自 a16z New Media 团队的 David Booth 与 Dhruv Gupta，共同探讨他们刚刚推出的全新项目——Cosign 奖项（Cosign Awards）。 回顾走入 2026 年底的这一年，行业环境发生了翻天覆地的变化。David Booth 坦言，长久以来在公众视野中享受掌声与聚光灯的几乎总是创始人、CEO 这些台前的明星人物，但实际上各个业务条线中往往隐藏着少数几位发挥决定性作用的一线操盘手（Operators）。正因如此，Cosign 奖项应运而生：它不仅旨在发掘全年在幕后产生巨大影响力的优秀操盘手、甄选出具有极高人才密度（Talent Density）的潜力企业，还要表彰改变行业格局的标杆产品。评选结合了去中心化的公众提名与高水准的专家评审机制，让那些深藏在公司内部却鲜为外界所知的杰出人才走向台前。 Dhruv Gupta 分享了他与 MTS 之间的真实经历，以此阐释“背书”的重量。作为 Cosign 系统的主要搭建者之一，Dhruv 回忆起自己初到旧金山的经历，正是因为 MTS 的首席执行官 Brent 为他做出了某种形式的背书（Co-signed），他才得以真正迈入整个初创生态圈。Sophia 对此深有共鸣，指出在现代职场中，最宝贵的往往不是冷冰冰的过往履历或上一家公司的名字，而是某个备受信任的人发自内心地断言“这个人值得雇佣、这个人在其专业领域极为出色”。Cosign 实际上是在将这种稀缺的人格背书转化为一种可规模化复制的网络机制。 - 正如 David Booth 在阐述背书的本质时所言：“One way you could say is like if you were willing to introduce this guy to somebody that you deeply respect that's incredibly busy, most people can't get to, but you say, actually, this is worth your time. That's the coastline.”（如果有一位极其忙碌、寻常人根本约不到且备受你尊敬的大佬，而你愿意把某个人引荐给他并告诉他‘这绝对值得你花时间’，这就是背书的真正力量。） - Dhruv Gupta 谈到创办奖项的初衷：“With anyone being able to nominate, anyone being able to vote, we hope that a lot of operators that are under-recognized by the public but have great recognition inside their company are able to kind of take that step to being well-known in the public for the great work they're doing as well.”（通过赋予所有人提名和投票的权利，我们希望让那些在公众面前被严重低估、但在公司内部备受认可的操盘手，能够因为他们所做的卓越工作而在行业中名正言顺地获得知名度。） ```mermaid flowchart TD accTitle: 传统招聘与基于 Agent 及信誉背书网络的匹配流程对比 accDescr: 对比传统冷启动简历投递的低效流程与基于个人背书和AI Agent撮合的高效意图重合模式 subgraph Legacy["传统求职模式"] L1["静态简历投递"] --> L2["海量冷邮件 / 招聘板"] L2 --> L3["HR 人工筛选 (噪音大、信任度低)"] end subgraph Cosign["Cosign 信任网络与 Agent 匹配"] C1["业内熟人 / 领袖背书 (Co-sign)"] --> C2["真实人才声誉图谱"] C2 --> C3["个人 AI Agent"] C4["企业 AI Agent"] --> C3 C3 -->|"意图重合 (Coincidence of Intent)"| C5["精准推荐至招聘决策者"] end ``` ### `05:00` 细分赛道映射、意图重合与“扩大运气的接触面积” Cosign 奖项的评选架构被梳理为三大支柱：人物（People）、企业（Companies）与产品（Products）。在人物维度上，不仅涵盖了工程师、全职员工、产品设计、AI 研究员、品牌传播与增长等传统岗位，还别出心裁地设立了公共知识分子（Public Intellectuals，已有 65 位被提名人）、年度播客主（Podcaster of the Year，已有 76 位被提名人）以及发帖达人（Poster of the Year）等紧跟互联网文化的特殊门类。在现场的笑谈中，Sophia 得知自己作为 MTS 主持人早已名列播客榜单。目前角逐最激烈的当属“明日之星”（Rising Stars）奖项，提名人数已超 135 人，其中既有刚拿到种子轮或 A 轮融资且潜力巨大的早期创始人，也有完全隐于幕后但被同行极力推崇的新锐。 企业维度的奖项除了按行业赛道划分外，更重点考察公司的“人才密度”。平台尤其鼓励员工自发为自家公司投票，因为内部员工的认可本身就是最坚挺的雇主信誉信号。当探讨这些数据沉淀后将赋能何方时，Dhruv Gupta 强调 Cosign 的终极目标是成为初创生态系统的“真相之源”（Source of Truth），无论未来是人类、企业还是 AI 智能体提问（例如“纽约做生物技术最酷的人是谁”或“最早发掘 Sophia Dew 的伯乐是谁”），系统都能直接给出高置信度的答案。 谈到人生的机遇，David Booth 引用了史蒂夫·乔布斯（Steve Jobs）的名言，指出职业生涯中的很多突破看似偶然，本质上却取决于如何“置身于运气的路径之上”。Cosign 想要做的，就是将这种偶遇系统化，促成“意图的重合”（Coincidence of Intent）。让原本素昧平生但抱负相通、能力互补的设计师获奖者与工程师获奖者在图谱中彼此发现，从而催生出全新的伟大事业。Sophia 赞同地指出，传统的学历筛选与招聘路径早已跟不上当下野蛮生长的创业节奏，依托社交图谱与深度声誉的背书网络，正成为新一代雄心勃勃的年轻人改变命运的关键跳板。 - Dhruv Gupta 描绘了该网络对生态的全面开放性：“This should be where you go to get any question you have answered, whether that's like, who in New York is building the coolest thing in biotech? Or like, who was like the first person who discovered Sophia Duke? All of those questions should be able to be answered from Cosign.”（无论是想了解纽约谁在构建最酷的生物技术产品，还是想知道最初是谁发掘了行业新秀，Cosign 都应当成为解答这些疑问的核心据点。） - David Booth 阐述了构建该平台的核心哲学：“How do you make two people who have the same hopes, dreams, and complementary skills come into each other's orbits soon? And the whole awards is really part of this, like maybe a winner of a design and a winner of an engineering award who otherwise never would have found each other have a better chance of doing that now.”（如何让两个拥有相同抱负、相同梦想且技能互补的人尽快进入彼此的轨道？整个奖项本身就是这个大蓝图的一部分，比如一位设计奖得主和一位工程奖得主，原本可能一生都不会相遇，但现在他们有了更大的机会走到一起并肩作战。） ### `10:00` 从冰冷的投递到信任背书：摆脱金手铐与职业冷启动 建立社交媒体影响力来获取职业机会往往门槛极高，而通过熟人的直接背书（vouch）则是一条更为精准且高效的捷径。Dhruv Gupta 分享了自己在宾夕法尼亚大学（Penn）读大四时的亲身经历：在大二和大三的暑期实习中，他走的是最传统的金融路线，先后在摩根士丹利（Morgan Stanley）和简街（Jane Street）工作。当时为了争取这些岗位，他每天都在求职平台 Handshake 上机械地海投，整个体验令人窒息且枯燥至极。直到后来他在 Polymarket 所做的工作被 Brent 关注到，Brent 认可他的潜力并力邀他加入 MTS，才真正为他打开了进入科技圈的大门。Dhruv 坚信，每个人在职业生涯中都需要一个愿意为自己押注的伯乐，而 Cosign 的核心诉求正是让这种信任背书系统化，打破传统招聘中盲目海投的恶性循环。 与会者随后探讨了现代职场中普遍存在的“金手铐”现象：许多极具天赋的人被困在薪水丰厚但缺乏激情的传统岗位上，甚至不知道外部世界还有哪些更适合自己的可能。如果能让这些人才去做自己真正热爱且独一无二的事情，去解决具体行业中的核心难题，世界将被真正改变。David Booth 进一步引申指出，改变世界的往往不是某一家特定的公司本身，而是一群志同道合的人聚在一起工作、建立深厚连结、并最终共同成为未来联合创始人的过程。这正是“你就是与你相处时间最多的五个人之和”的体现——无论是在美国再工业化、发掘稀缺关键矿产还是其他关键使命上，当人才主动筛选进入高度契合的文化圈子时，就会发现职业岗位虽然短暂，但在沿途结识并相互背书的伙伴却能够长久地支撑彼此去打造下一个事业。 Dhruv Gupta 极其尖锐地对比了海投与信任推荐两种招聘方式的未来差异，他指出：“I think if we move more and more towards that as a means for hiring, instead of like cold apply button mashing on LinkedIn or handshake, it'll be like a better world for everyone.” ```mermaid flowchart TD accTitle: 招聘模式对比与人才释放路径 accDescr: 对比传统平台冰冷的海投模式与基于Cosign信任背书的流动模式 subgraph Traditional ["传统求职模式"] A["求职者在 Handshake/LinkedIn"] -->|"无尽机械海投"| B["被困在金手铐岗位"] B --> C["缺乏激情的重复性工作"] end subgraph CosignModel ["Cosign 背书模式"] D["同行/导师观察到真实潜力"] -->|"建立信任背书 (Cosign)"| E["精准引荐进入前沿团队"] E --> F["聚集高密度人才圈子"] F --> G["共同孵化未来项目与公司"] end ``` --- ### `13:14` Cosign 的三层架构：从维基式主页、策展清单到深度背书 对于渴望转换赛道的年轻优秀人才而言，另一个巨大障碍在于信息不对称：巨头科技公司的名字无处不在，而潜力巨大、正在解决硬核问题的小型初创团队却极难被发现。尽管像 a16z 投资合伙人 Katie Kirsch 在 LinkedIn 上发布的招聘动态已经成为业内寻找被投公司机会的重要风向标，但 Cosign 正在通过更底层的机制重塑这一过程。David 阐述了 Cosign 的三层产品逻辑： 第一层是类似维基百科（Wikipedia）的结构化档案层，平台已经为 40,000 到 50,000 名行业从业者搭建了基础 Profile，用户可以顺着公司、投资人和从业者的节点自由双向跳转探索；第二层是清单（Lists）机制，属于轻量级的行业策展。目前已有约 400 人在平台上创建了公司和人物清单，例如知名商业分析通讯《Not Boring》的作者 Packy McCormick 就在其清单中列出了 6 家他深入研究过且高度认可的垂直整合型公司。此外，平台上还收录了 Interact 学者、Thiel 奖学金获得者等群体的历史清单，使用户能够清晰追踪 2014 年以来的顶尖人才如今都分散在何处。第三层则是皇冠上的明珠——极为郑重其事的“Cosign”（共同背书）。它代表着极高的个人声誉背书，传递的信号是“无论这个人去哪里我都愿意与他共事”或者“他改变了我的职业轨迹”。 David Booth 总结了平台目前构建的基础数据骨架：“I think of cosign in sort of these three layers. One is, it's almost like Wikipedia style. You've scaffolded out profiles for 40,000, 50,000 people across the site now.” - **底层主页规模**：目前系统已自动构建并整理了 4 万至 5 万个从业者主页，实现人、机构与投资方的高效穿透。 - **清单功能（Lists）**：支持无限滚动浏览（`cosign.co/list`），汇集了约 400 位投资人、GP 及行业观察者（如 Packy McCormick）整理的垂直领域推荐清单。 - **群体轨迹比对**：支持多维度交集查询，例如比对 2014 届 Interact Fellows 或 Thiel Fellows 的当前流向。 - **发掘未被发现的天才**：David 提到，产品最激动人心的愿景是给全球范围内的“下一个 50,000 个 Dhruv”提供被看见的机会。 --- ### `16:47` 打造 AI Agent 原生网络：打破 LinkedIn 围墙与构建人才数据层 在讨论到发掘全球未被开发的潜力时，嘉宾们分享了一个生动的案例：MTS 在招募工程师时，曾收到一位来自瑞典的 17 岁高中生的自荐。尽管当时受限于条件未能直接录用他，但团队一直在其他职业轨道上为他提供帮助。这类散落全球的天才往往受限于地域和传统认知壁垒，无法直接获取美国的顶尖机会，而 AI Agent 的介入将彻底改变隐藏在创投生态背后的数据层流动机制。 Dhruv 直言，像 LinkedIn 这样的传统巨头有着极强的围墙花园属性，永远不可能对外开放模型上下文协议（MCP）来让外部 AI Agent 自由检索其图谱，导致用户在海量数据面前依然寸步难行。Dhruv 提到，自己夏天在纽约时曾想寻找能够加入的顶尖初创公司，不论是 Google Gemini 还是 ChatGPT，给出的答案都相当空洞乏味，原因就是缺少真正反映人才与团队高质量关联的数据层。Cosign 旨在填补这一空白，让用户在完善档案、整理清单和互相背书的过程中不断丰富关系图谱。 为了让 AI 代理真正理解人才供需，Cosign 正在构建一套 Agent 原生的使用体验。用户不仅可以使用站内的 Ask Cosign 功能，还可以通过 iMessage Agent 进行即时对话，甚至可以通过个人 MCP（Model Context Protocol）将自己的 Muse、GrokBot、Instinct 等 AI 助手直接接入 Cosign 平台。David 随后指出，在人才市场获取超额收益（Alpha）就像量化基金做投资一样，要么利用大家都能看到的公开数据进行算力挖掘，要么就是掌握更深层、尚未被公开标记的真实关系链与信任数据。 Dhruv Gupta 指出了传统职场社交平台在开放性上的根本局限：“So I think that one thing that... like LinkedIn will never do, is introduce an MCP that allows you to have vast search access over the graph.” ```mermaid flowchart LR accTitle: Cosign Agent 原生数据架构与接入方式 accDescr: 展示用户互动丰富关系图谱，并通过多样化的Agent接口向外部AI模型开放 subgraph DataEnrichment ["数据图谱层"] U["用户行为与背书"] -->|"贡献 Profile / Lists / Cosign"| G["Cosign 关系图谱"] G -->|"沉淀隐性关系与意图数据"| DL["结构化人才数据层"] end subgraph AccessLayer ["Agent 原生访问接口"] DL -->|"平台内置"| AC["Ask Cosign"] DL -->|"移动端即时交互"| IM["iMessage Agent"] DL -->|"开放协议"| MCP["Personal MCP Server"] end subgraph ExternalAgents ["外部集成生态"] MCP --> Muse["Muse"] MCP --> Grok["GrokBot"] MCP --> Instinct["Instinct"] end ``` ### `20:00` 个人智能体代理与“意图契合”的网络对接 随着底层模型的演进，公开且商品化的人才数据将能够被更深度地分析，但更引人深思的是当智能体（Agent）能够代表各自的人类主人相互谈判时会发生什么。近期个人 Agent 的爆发式发展速度令人咋舌，短短一两个月内相关应用就积累了千万级用户。个人 Agent 对人类用户的了解可以极其深入，知晓其所有职业抱负、升职疑问或技能提升计划。然而，现阶段 Agent 仍然缺乏一种直接与发掘人才的人建立联系的机制；当前它所能做到的极限，往往不过是在收件箱里草拟一封电子邮件发给对方，等待对方的 Agent 提取并回复。近期在客户成功（Customer Success）领域已经出现了早期 Agent 间通信协议（Agent-to-Agent Protocol）的探索案例，例如租车大公司与个人用户的 Agent 之间可以直接完成协商交互。 沿着这种思路延伸到职业网络领域，如果一个人正在寻找工作，最理想的状态是求职者的 Agent 能够直接对接公司招聘主管的 Agent，或者在两个人想要结识时，先由双方的 Agent 进行第一轮沟通。Agent 之间可以预先核对彼此的共同目标，确认是否存在真正的“意图契合”（coincidence of intent）；一旦确认双方意图匹配，Agent 再促成人类之间的实际连结。为了实践这一构想，团队目前正在加紧构建专属入口（cosign.co/agents），支持用户连接其 MCP（Model Context Protocol）、通过发短信与 Agent 交流，或是直接在内置 UI 中进行“Ask Cosign”对话。 现在的职业网络普遍存在信号失真的严重缺陷。正如团队所指出的，在现有平台上，Sam Altman 的收件箱和普通人乃至其 14 岁妹妹的收件箱并没有本质机制上的区别，所有人都在接收杂乱无章的冷邮件，而这种机制显然是不合理的。Cosign 的核心信念之一正是最大程度地保全高价值社交信号，确保用户只将宝贵时间花在真正该见的人身上，而不必在大量无意义的低效连接中耗费精力。 ```mermaid flowchart TD accTitle: Agent间意图契合与撮合流程 accDescr: 展示个人Agent如何与招聘主管Agent沟通，只有在达成意图契合后才连接人类。 subgraph S1["意图与档案理解"] U["个人用户"] -->|"职业诉求与技能"| PA["个人 Agent"] C["招聘主管 / 合作方"] -->|"招聘需求与组织目标"| CA["企业 / 对手方 Agent"] end subgraph S2["Agent-to-Agent 协议层"] PA <-->|"谈判 / 诉求对齐"| CA CA --> DET{"是否存在意图契合？<br/>(Coincidence of Intent)"} end subgraph S3["人类连接与筛选"] DET -->|"否"| DROP["过滤低质骚扰与冷邮件"] DET -->|"是"| MEET["促成双方人类直接沟通建联"] end ``` - David Booth 或 Dhruv Gupta 在剖析现有收件箱机制的痛点时强调：“Yeah, I mean, on networks right now, the inbox works the same way for Sam Altman as it does for me, as for my 14-year-old little sister.” - 关于筛选与维护高价值连结的核心愿景，团队补充道：“And I think that that's one of our core beliefs as well, is that we really want to be able to preserve that signal and make sure that you are meeting the people who you should be meeting.” --- ### `22:21` Cosign Awards 奖项规划与产品快速迭代 针对未来几个月的发展规划，Cosign 当前的首要工作是推进第一届 Cosign Awards 的评选落地。该奖项的提名与投票将持续开放至 11 月底，留给社区的参与时间大约还有三周。Cosign Awards 最具特色的地方在于，经评委与社区共同评选出的各个类目获奖者，将受邀齐聚一堂。团队认为在当下的职场环境中，优秀人才并不缺少一个额外的奖杯或证书，真正宝贵的是能够与其他获奖顶尖从业者面对面建立深层关系的难得契机。 在产品研发方面，由于 Cosign 团队处于高强度的快速迭代状态中，甚至很难预知几个月后产品会演化成什么具体形态，一切完全取决于用户的真实反馈。近期团队反复收到的最核心反馈是入驻与上手引导（onboarding）必须变得更加丝滑流畅，这也是团队深入了解如何为用户解决问题的关键切入点。主持人 Sophia Dew 在节目收尾时表示，很荣幸见证 Cosign 从最初发布到如今推出 Awards 的演进，并与嘉宾约定在一个月后重回 MTS 节目现场公布最终的获奖者名单。 - 团队在谈到奖项对从业者的核心价值时指出：“So I think that people don't need another trophy or another certificate. The opportunity to build relationships with the other category winners is the real thing I'm honestly the most excited about.” - 团队在面对产品路线图的快速迭代时坦言：“And I can't answer what it's going to be a week from now, much less a few months from now. All you know is that you're busy. What we can say is that the product changes rapidly with feedback.” - Cosign Awards 提名与投票截止日期为 11 月底，各细分赛道的优胜者将由评审团参与最终裁定，并受邀参加专属的线下闭门交流。 - 团队近期产品优化的重中之重是重构 Onboarding 流程，消除用户在初次使用与构建个人声誉网络时的摩擦阻力。 ## Notable quotes > `20:00` Yeah, I mean, on networks right now, the inbox works the same way for Sam Altman as it does for me, as for my 14-year-old little sister. — David Booth > `22:21` So I think that people don't need another trophy or another certificate. The opportunity to build relationships with the other category winners is the real thing I'm honestly the most excited about. — David Booth > `16:47` So I think that one thing that... like LinkedIn will never do, is introduce an MCP that allows you to have vast search access over the graph. — Dhruv Gupta ## Mentioned Cosign, Cosign Awards, Sophia Dew, David Booth, Dhruv Gupta, a16z, a16z New Media, MTS, Brent, LinkedIn, Handshake, Morgan Stanley, Jane Street, Polymarket, Packy McCormick, Not Boring, Katie Kirsch, Thiel Fellowship, Interact Fellowship, Google Gemini, ChatGPT, Model Context Protocol (MCP), Muse, GrokBot, Instinct, Sam Altman, Steve Jobs ## Video description What if finding your next job, co-founder, or collaborator depended less on a resume and more on who is willing to vouch for you? Cosign is building a professional network around those signals, with the longer-term goal of giving AI agents access to the relationship and reputation data that traditional networks keep locked away. a16z New Media’s David Booth and Dhruv Gupta join MTS to introduce the Cosign Awards, which recognize standout operators, companies, and products based on nominations from the people who know their work. They also unpack the bigger idea behind Cosign: mapping the people and relationships that rarely show up in public data, then making that graph accessible to both humans and agents. Looking ahead, they imagine professional networking where your personal agent can talk directly to a company’s agent, determine whether there’s a potential fit, and only bring the opportunity to you when there’s a genuine “coincidence of intent.” [Source](https://mgln.ai/e/1344/afp-848985-injected.calisto.simplecastaudio.com/3f86df7b-51c6-4101-88a2-550dba782de8/episodes/fcd03bbe-16a1-453c-a1c0-69adfb573df8/audio/128/default.mp3?aid=rss_feed&awCollectionId=3f86df7b-51c6-4101-88a2-550dba782de8&awEpisodeId=fcd03bbe-16a1-453c-a1c0-69adfb573df8&feed=JGE3yC0V)

[20:00] David Booth: Yeah, I mean, on networks right now, the inbox works the same way for Sam Altman as it does for me, as for my 14-year-old little sister. [22:21] David Booth: So I think that people don't need another trophy or another certificate. The opportunity to build relationships with the other category winners is the real thing I'm honestly the most excited about. [16:47] Dhruv Gupta: So I think that one thing that... like LinkedIn will never do, is introduce an MCP that allows you to have vast search access over the graph.

[20:00] Dhruv Gupta: On networks right now, the inbox works the same way for Sam Altman as it does for me, as for my 14-year-old little sister. [17:15] Dhruv Gupta: So I think that one thing that like LinkedIn will never do, is introduce an MCP that allows you to have vast search access over the graph.

[0:00] What if I'm looking for a job and in fact like my agent should be able to connect with the company's agent and do a first pass of whether it's worth elevating me to the hiring manager or maybe it's true of and I are looking to connect and it's our agents are having a first conversation. Whether it's like what are our mutual goals, what is the coincidence of intents and is there one here where if there is then the agents can connect us. The biggest opportunities in your career often come down to one person willing to vouch for you. when you turn that signal into a network. In this episode of MTS, Sophia sits down with A16Z New Media's David Booth and Dhruv Gupta to discuss Cosign and the new Cosign Awards, recognizing the operators behind some of the most successful companies. They also explore a bigger idea, a professional network built around reputation and trusted relationships, where AI agents could eventually connect people and companies without the noise of job boards and cold outreach. Falls, your surface area for luck. Hello, everyone. Welcome back to MTS Today. We are joined live by two guests. We've got David Booth here and Dhruv Gupta. They're both from A16Z's new media team, and they just launched Cosign Awards. So excited to have you guys on MTS Today to talk a little bit about it. Excited to be here. Dad, do you want to kick us off with what Cosmetic are? Yeah. Yeah, let's kick it off. We did. I mean, we're coming in towards the end of 2026. Um... And it's been crazy. It's been a crazy year. One of the things we keep seeing again and again is that sometimes, you know, the people that get the recognition in the public, the founders, the CEOs, the face of the company. But more and more, in this year in particular, I think going forward, there's across all of the functions, there's like one or two or, you know, there's some smaller number of people who are just the operators who are like having the most impact. anyone across it. So Cosone Awards is honestly, the first idea was, well, can we shine a spotlight on who are the people that had the most impact across the year? Which are the companies that have the highest talent density that have the most impact on the world, but also could have the most impact on your career? Which are the products that had the most impact throughout the year? And so bring that together and to do this combination of the incredible opportunity for people to nominate those that they see that someone might not see. And also for people who, you know, we've selected this pool of judges, we'll talk about some more who have this visibility spanning the industry and can talk about, you know, what they're seeing at the high level in the meta trends. And hopefully we pull this together into some really, really interesting people to celebrate, companies to celebrate. This is pretty cool. I would say one of the big things that I saw from Cosign Awards is that who are some of the top people when it comes to Twitter growth. But that information isn't known across you guys, or you guys know some information across other companies that we don't quite know. So what's the big goal here with Cosign Awards? Maybe, Drew, I know you played a really impactful role in actually building this system. So tell me more about that. Thank you. Yeah, I think the Cosign Awards, our main goal is to... give people the chance to celebrate everyone at their company that they think deserves it. I remember like Being at MTS, that was really my first experience in San Francisco. I think that I really could only get into the startup ecosystem as a whole because Brent, the CEO of MTS, co-signed me in one way. And I think co-sign awards are a way for us to also try and give all those people who are doing great jobs that visibility. With anyone being able to nominate, anyone being able to vote, we hope that a lot of operators that are under-recognized by the public but have great recognition inside their company are able to kind of take that step to being well-known in the public for the great work they're doing as well. Yeah. You brought up an important point right there around Brent co-signed you. Oftentimes, it's not the resume that's the impressive piece or your past job, but it's the person that I trust who can vouch for you. It's that connection. That's that person who can actually co-sign and prove that this person is worthy or this person should get hired or this person is phenomenal at what they do. And a little bit what you guys are doing is you're just now allowing that to happen on a massive scale. Yeah. What are some of these things, awards that you could potentially win here. There's, I mean, firstly, exactly that. One way you could say is like if you were willing to introduce this guy to somebody that you deeply respect that's incredibly busy, most people can't get to, but you say, actually, this is worth your time. That's the coastline. In the same way here, I'd say across of [5:00] What we tried to do is we tried to break, and that's not perfect, we couldn't get as granular as you'd like to, but break down across those three categories, the people, companies, and products. What's the most logical way to think about each? For people, a lot of them are role-based. So there's things like engineers, FTE, product design, there's things like AI researchers, comms growth. There's a couple that are less traditional on the org chart, like it's a category for public intellectuals with 65 nominees. There's one for podcaster of the year, 76 nominees. The one that- Well maybe we can end up on it. Well, I was gonna say, I mean- I think you are on. There's a video, yeah. Sweet! If all the MTS hosts out there already, then we're doing something wrong. Poster of the year, I mean, any, you know, a lot of the audience here will be pretty ex-native anyway, so they'll see a lot of familiar faces in there. There's ones I'm personally very excited about. I think the rising stars is the most populated category. We could share this or otherwise people should head to cosine.co/rewards. but there's 135 so far and there's some really interesting people. There's a mix of like, I see founders who are, you know, series C through series A, but still a huge ramp ahead of them. I see people who are like, you know, some are better known and some I've never heard of ever in my life. But, you know, I sure know that if someone out there thinks they're a rising star, then that is a cosine. And that's something that I'm going to take note of and pay the interest to. Yeah. And similarly for companies, we have a lot of awards that you can win by industry, but also general awards for things like talent density. We want like employees especially to be voting for their own companies because that tells us that, you know, this is a place that they like to work and probably a place that like other people would also like to work. That's really interesting too. So the people that would benefit from this Who's this for? What is all this information for? uh, I think it's for everyone. I think we want Cosign to be the source of truth for the startup ecosystem. This should be where you go to get any question you have answered, whether that's like, who in New York is building the coolest thing in biotech? Or like, who... Who was like the first person who discovered Sophia Duke? All of those questions should be able to be answered from Cosign. And I expect that like people will be interested in it. Companies will be interested in it. Agents will be interested in it. We're really building this for everyone. Yeah. Agents will be interested. Interesting. I mean, personally, I've spent most of, if not all of my career in pursuit of at this point, is the idea that a huge amount of what happens around us comes down to what feels like coincidence, what feels like coincidence. you know, just meeting the right co-founder at just the right time or, you know, conversation with a friend of a friend who introduces you to the person that offers you a breakout job or the cold email you sent. It's like what happened to create that coincidence? There is the, as tweeted about this earlier today, but the Steve Jobs quote I've always loved, that you can't plan to meet the person who changes your life. But you can certainly put yourself in the path of luck and you can know that, you know, when that person appears, you can grab it with both hands and run with it. one way I've started reframing Cosign is yes, I hope that it can be a source of truth for the tech ecosystem. I hope there can be a place that you can go and ask questions that you can't ask anywhere else. But like, and it's almost easiest form. I hope it's a place where you can increase your service area for luck. It's a place where you can go and you know that if you build a profile out or if you browse the graph and you sort of you teach cosine, you talk to, there is an agent, we're talking about that. You ask cosine questions, then it's going to help you put yourself in the path of luck. It's going to help people find you. It's going to help your future co-founder find you and realize that you might be into the same things. This idea of a coincidence of intent. How do you make two people who have the same hopes, dreams, and complementary skills come into each other's orbits soon? And the whole awards is really part of this, like maybe a winner of a design and a winner of an engineering award who otherwise never would have found each other have a better chance of doing that now. And if that happens, then they build something together, then we're all better off for it. Yeah, it's pretty interesting because it feels like the legacy systems for maybe how you would traditionally get a job are outdated. Just the idea, even just going to college, a lot of young people are like, do I need a degree to get my dream job? ambitious people who just show up with a big dream. And for a lot of people, social media is that outlet that allows them to actually get an opportunity. [10:00] land them their first job, open up doors. But It's not that clear of a path. It's really hard to build up a social media following and try to create those opportunities for yourself. So it feels like cosine is almost this shortcut. You know, if you already have some of these people, you can get some of those opportunities that maybe social media traditionally could give you, but just get them through people you already know who can vouch for you. Yeah. Yeah. I mean, I'm a senior at Penn right now. In my career, in my summer internship, sophomore and junior year, I went down the very traditional path of Morgan Stanley and Jane Street finance stuff. Yeah. And to get those jobs, I was just like cold applying on Handshake every day checking new jobs. It was like, Horrible. It was not a fun experience at all. My first experience in tech, as I mentioned earlier, was through Brent, seeing some of the work I'd done at Polymarket before and recognizing that I would be able to contribute at MTS. And him taking that chance on me really allowed me to get into the tech world. I think that Everyone will have someone like that who can take that chance on them. And we want Cosign to be the place where that happens. And you can find the people who the people you trust are cosigning and like, think about giving them jobs. I think if we move more and more towards that as a means for hiring, instead of like cold apply button mashing on LinkedIn or handshake, it'll be like a better world for everyone. I completely agree. I think there's one piece that you just alluded to, which is the person piece, where now you can actually find people and opportunities because there's a person that vouched for you. But on the other end, there's also this company piece, I see it here, David, that you're pulling up, One thing I think that's interesting here is if you can get all the people who maybe have a traditional job that they don't love. Maybe some people love their traditional jobs, so keep them wherever you're excited there. But there's a lot of amazing talent stuck with golden handcuffs who, for whatever reason, don't know what else exists. And I have a fundamental belief that if you get all of these people to genuinely do what they're passionate about and what they're exceptional and one of one in, in an industry or solving a problem that a specific company that's how you change the world. Just get everyone to work on the thing that they were meant to be doing. It's yes, and it's not even that company itself. which needs to change the world, it's the fact that a lot of people who think that way go work there together, and they build their own relationships, and they meet their own future co-founders. And it's kind of, it's an extension of the idea that you are the average of the five people you spend the time with. So if you want to, you desperately want to help, you know, re-industrialize America, or, you know, uncover new sources of essential minerals, or, you know, whichever one of the thousands missions that people are working on out there, if you self-select into a culture of people who are working on that mission, you will find that it's not even just the, there's this line of careers along, roles are short, but it's the people you made along the way that you can build the next thing with. And co-sign when you're there with them so that you can keep tabs on them. Yeah, completely agree. I would say for a lot of this young talent too, or talent that just, exceptional talent that maybe wants to make a career change, it can be really hard to find what are those companies that they should go towards. You hear about the big companies because you consistently hear their names again and again, but you don't hear about some of these smaller startups. And so I know that's something that you guys have been very focused on. I know famously Katie Kirsch with her LinkedIn posts is one of the best ways to find out about new types of jobs within the A16Z Portco world. But tell me what Cosign's role is here. There's, I mean, if you think, again, Awards Live Now, they're kind of community-generated lists. too. There are lists all across. I mean, it's one of the core things that we're playing with. I think of cosign in sort of these three layers. One is, it's almost like Wikipedia style. You've scaffolded out profiles for 40,000, 50,000 people across the site now. You can click through, you can rabbit hole up and down the companies or investors, back to people. The next layer up is lists. It's a way of doing a lightweight curation. So you can go to many, I think about 400 people at this point have made lists of companies. Paki McCormick, Not Boring, is one six companies listed there, which he and his time that he takes to write and research thinks are really interesting, vertically integrated, you know, companies that somebody who reads not boring might be interesting going to work at. But then that equivalent across of however many like angel investors and GPs at great funds, these are the companies that if you're looking for something, you should go and join one of these. If you go and sort of dig into, there's this tab, cosine.co/list. [15:00] scroll down, It's an infinite scroll and you go through all of these different subsets of That's the company ones, but also people. What if you could look at every different individual list? There's sort of the Interact fellows, the Teal fellows, the different lists that have been created over time and see where they intersect. See where the Interact fellows of 2014 are now is the one that has to be on my screen in front of mind. of that, then the cherry on top of that is the cosine. You talk about that. It's the very intentional given from me to you to say, this is somebody that I would work with anywhere, someone that I think is going to be great and the world doesn't know it yet, or someone that changed my career and helped put me in business. And honestly, I do want to say as well, this is very, very early in this product development. There's a lot more that we're testing, iterating on live. I would encourage for us. You're like, you know, it's still some rough edges on the experience. Cosines, particularly, we're going to be leaning in on in the future. I think one of the most important things, Drew is very, very passionate about this, given his story and Brent and things as well, is I think there is like one of the most interesting things this could do is be that opportunity for the next 50,000 Drews. He has no equal, but yeah, there's another 50,000. How many thousand people out there? just have incredible potential and they haven't been discovered yet. And they haven't been discovered yet. Yeah, 100%. Especially thinking about that when it comes to a global, you know, a global level, because it's not just in the United States. There's probably some amazing, exceptional people in other countries that can never find out about the opportunities here in the U.S. And now with social media, with ability to network across borders. Who knows what's going to be possible? We can tap into everyone's untapped potential. you know, You know, it's incredible. Back when I was at MTS, when we were trying to hire engineers, there was a 17 year old kid from Sweden who reached out to me about that role. And even though we weren't able to have him work with us at MTS, I've been able to help him in other ways in his career. And I'm sure that there's so many other thousands of young people like that around the world. of agents to then parse those lists and get information might create the next data layer for these hidden networks behind the startup ecosystem. Yeah. So I think that one thing that... like LinkedIn will never do, is introduce an MCP that allows you to have vast search access over the graph. So even though there are so many thousands, hundreds of thousands of accounts on LinkedIn, your ability to find the people that you need is limited. And the same thing for companies. I think I remember back when I was in New York this summer, I was like, I wanted to start like meeting people who I thought I might be able to work for one day. So I Googled like, who are the best startups in New York? And Gemini gave me like a pretty bad answer. I switched to ChachiBT. It also gave me a pretty bad answer. I think that that like whole data layer of information is kind of, missing right now. And our hope is that with Cosign, with people contributing to their profiles, contributing to lists, enriching our graph even more as agents are able to to like search through that we'll be able to get like much better answers to those questions much better answers to those questions personalized to you because like as you are interacting with the platform we're getting a better understanding of what you're looking for what you would want to do what answers would like make you the happiest um And we want you to also be able to access these agents anywhere. We have like our on platform ask cosine feature, but we also have an iMessage agent. We also have a personal MCP that lets you connect your Muse, your GrokBot, your Instinct, etc. to the platform. And we're going to keep on building more and more of those types of integrations to make this truly the first kind of agent native professional networking platform. That's so interesting to me because a lot of this information right now is completely gated away. You can't actually parse it. There's the... There's so many fun layers to this, and I'm going to start with a call out to people who are thinking about this kind of stuff. DMs are open, and I want to chat. But there is a lot of what Drew was talking about, rightly, is about the search problem. And search is, I think, still a problem, but it's getting easier. There is a new... Yeah, extra is phenomenal. There's so many different databases you can search across, you've got more powerful models. I think that there's kind of two ways to find, to get alpha in talent. One of them is like a quant fund. You take the same public information that every other fund has. [20:00] but you can analyze it better. And I think that as models get better, you're going to be able to analyze the public commoditized talent data. And Uh, The thing that I've been thinking a lot about is what does it mean when agents can kind of negotiate with each other on behalf of their owners? So I've been following along so closely with the personal agent wave of the last month. It seems insane that it's only a month or two. Muse has hit 10 million users now, I think. There's this line, I realized that your personal agent could know you extremely well. It can know all of what you want to achieve. You're asking questions about how to get promotions and how to find new jobs and how to develop skills. But it still doesn't have a way to actually connect with the people that are finding you. The best it can do is draft an email in your inbox and send it to someone else's inbox, where maybe an agent picks it up and sends back in. So we realized that recently there was some very early showing of agent-to-agent communication and customer success was one. was one of the first to introduce their agent-to-agent protocol with Muse and Instinct. And their customers who might be big companies, like a car rental company, say, and the personal user can just negotiate directly with the agent on the other side. And then we started thinking, you go down that rabbit hole, you realize that, well, what if I'm looking for a job? And in fact, like our agents should connect with, my agent should be able to connect with the company's agent hiring manager or maybe it's driven I looking to connect and it's our agents are having a first conversation whether it's like what are our mutual goals what is the the coincidence of intense and is there one here where if there is then the agents can connect us so there's a lot of fun to be had here we're basically at the moment prioritizing building at cosine.co forward slash agents I don't know if we tweeted that out yet but people can go there and they can connect their MCP they can text the agent they can find ask cosine super powerful just conversational baked right into the UI. And there's a lot more coming. Yeah, I mean, on networks right now, The inbox works the same way for Sam Altman as it does for me, as for my 14-year-old little sister. And I probably shouldn't. And I think that that's one of our core beliefs as well, is that we really want to be able to preserve that signal and make sure that you are meeting the people who you should be meeting. And you don't have to spend time looking through a ton of people who you probably shouldn't be meeting. Definitely. One last thing for you guys, since I know we're kind of at time. sign what can people expect to see from you all in the next few months Well, we've got to close out these first-hand awards. That's a big one. Voting is open or nominations are open through the end of the month. So through the end of November. So we've got about three weeks left. Now, the coolest thing about this, by the way, is that everybody who is with the judges' participation selected as the winners of the categories we're going to invite to bring together. So I think that people don't need another trophy or another certificate. The opportunity to build relationships with the other category winners is the real thing I'm honestly the most excited about. So that's a big one for us. A lot of product coming. So much product coming. And I can't answer what it's going to be a week from now, much less a few months from now. All you know is that you're busy. What we can say is that the product changes rapidly with feedback. So the more that people can come and share, one of the biggest ones that we've been hearing over and over again is that the onboarding needs to be a lot smoother and a big opportunity to learn how we can help people. So there's a lot of changes coming there. Just hit us with your questions, and we're excited to be building more. Awesome. Well, thank you guys so much for joining us today on MTS, announcing the Cosign Awards. I had you on earlier when Cosign first launched, and so excited to have you on again to talk about the latest updates. But thank you so much for joining us. We'll have to come and announce the winners in a month. I can't wait. I can't wait. It'll be happening live on MTS. Maybe we'll get a little exclusive. It'll be fun. podcast. If you like this episode, be sure to like, comment, subscribe, leave us a rating or review and share it with your friends and family. For more episodes, go to YouTube, Apple Podcasts, and Spotify. Follow us on X at A16Z and subscribe to our Substack at a16z.substack.com. Thanks again for listening, and I'll see you in the next episode. This information is for educational purposes only and is not a recommendation to buy, hold, or sell any investment or financial product. Thank you. [25:00] Capital Management, LLC, A16Z, or any of its affiliates. Information is from sources deemed reliable on the date of publication, but A16Z does not guarantee its accuracy.