A16z Debuts Cosign to Replace Resumes With Peer Endorsements

a16z推出Cosign：以同行背书取代传统简历 随着人工智能用自动化申请淹没招聘渠道，可验证的同行确信成为在共识形成之前识别顶尖人才的唯一稀缺信号。 讨论嘉宾深入剖析了Cosign背后的构建逻辑。该平台旨在取代LinkedIn这类扁平且二元的人脉关系连接，转而提供细粒度、高确信度的信号，例如关键导师、经受过验证的协作者以及对潜力新人的看好。随着AI驱动的自动化用无尽的触达噪音淹没招聘渠道，真实的人类确信已成为整个科技生态中最稀缺的筛选资产。借鉴早期LinkedIn、AngelList以及类似Wikipedia的社区策展经验，嘉宾解释了结构化背书如何为那些缺乏公众知名度的早期工程员工等幕后从业者创造流动性。最后，他们认为，公开发挥社交资本去支持尚未被证明的建设者，将带来具有复利效应的长期优势，既能巩固持久的声誉，又能吸引未来的优秀人才。 a16z与生态从业者推出了去中心化声誉名录Cosign，旨在记录真实的同行背书、导师指导经历以及对初创企业潜力新秀的早期押注。 (0:53) 随着AI自主智能体用数千份自动生成的申请席卷招聘渠道，真实的人类确信与经验证的同行背书正成为选拔顶尖人才的首要稀缺筛选标准。 (34:25) 与LinkedIn的二元连接图谱不同，Cosign将声誉结构化为细粒度的基本单元——包括指引职业生涯的导师、实战战友以及值得关注的新星——从而让从业者的履历具备流动性。 (4:06) Cosign借鉴Wikipedia的模式化解社交网络的冷启动阻力，在用户认领主页前，根据公开网络公告、引用推文和股东名册记录预先填充经验证的个人资料。 (42:12)

As traditional university credentials lose their signaling monopoly, decentralized peer endorsements—or 'cosigns'—have become the primary mechanism for discovering non-consensus talent in technology. Erik Torenberg, David Booth, Josh Elman, and Olivia Moore argue that in an era of automated, AI-driven applicant volume, verifiable human conviction and granular reputation records are critical to overcoming market information asymmetry. Unbundling Credentials Through Public Proof-of-Work (0:53) The traditional bundle provided by universities—instruction, networking, and accreditation—is increasingly being replaced by public proof-of-work and peer attention. Erik Torenberg observes that within the technology sector, being followed on social platforms by figures such as Patrick Collison, Elon Musk, or Marc Andreessen provides far stronger career signaling than a degree from an elite institution like Stanford. This shift has enabled individuals without conventional pedigrees, such as commentator Dwarkesh Patel or Cursor co-founder Michael Truell, to gain rapid industry leverage based on public execution rather than decades spent climbing corporate ladders. Historically, this reputational transfer occurred through kingmaker endorsements, exemplified by Paul Graham publicly ranking Sam Altman alongside Steve Jobs in 2009. Staking social capital on rising, unproven talent creates an asymmetric dynamic: it bypasses traditional gatekeepers for the recipient while establishing valuable early dealflow and status for the endorser. AI Application Floods and the Scarcity of Conviction (34:25) As generative AI and autonomous browser agents drastically reduce the effort required to apply for jobs, talent evaluation faces severe noise. Olivia Moore notes that an informal open call on LinkedIn for portfolio startup roles attracted 2,000 Google Form submissions, warning that automated agents will soon inflate applicant pipelines to tens of thousands of auto-generated outreach messages. In an environment of near-infinite, low-effort interest, authentic human conviction becomes the only genuinely scarce filter. Simultaneously, critical contributors within breakout companies—such as an early 'engineer number four'—suffer from severe reputational opacity. Because public credit concentrates almost entirely on founders, high-performing operators lack structured, legible mechanisms to port their verified execution records into future fundraising rounds or career transitions. Resolving this friction requires moving from self-reported resumes to validated social proof of actual contributions. Replacing Binary Connections with Contextual Primitives (4:06) Incumbent professional networks flatten all working relationships into binary connections, which Josh Elman and Olivia Moore point out frequently leads to awkward exchanges where first-degree contacts admit they barely know one another. To restore high-signal communication, Cosign structures reputation around three specific primitives: influential career mentors, proven peers with whom an operator has survived high-pressure startup 'trenches,' and early speculative bets on emerging talent. To manage network timing mismatches without causing awkwardness, the speakers advocate combining public directories with private, mutual-intent signals. Torenberg highlights an SV Angel model where pre-cleared, double-opt-in interest created enduring professional collaborations. Allowing operators to privately register intent—such as expressing interest in backing or hiring someone only if they leave their current role—enables talent matching to occur frictionlessly the moment both parties align. The Compounding Economics of the Early Believer (48:09) Addressing the objection that talent identification is proprietary 'alpha' that investors should keep private, David Booth argues that the secrecy argument only applies over ultra-short horizons. Over a multi-year timeline, publicly staking reputation on an undiscovered builder cements the endorser's position as their foundational champion, creating durable loyalty that transactional, late-stage capital cannot duplicate. The panel points to Brian Chesky’s public letter during Airbnb’s IPO: despite raising billions from later institutional funds, Chesky singled out his earliest believers, Y Combinator and Paul Graham. Publicly demonstrating conviction in non-consensus talent yields compounding dividends, permanently binding founders to their earliest advocates while acting as an inbound magnet for future builders. The speakers concede that replacing LinkedIn is neither feasible nor necessary, given its deeply entrenched 20-year network effect among the massive cohort of the 'mildly happily employed.' Instead, Cosign is positioned as an open, verticalized directory tailored specifically to technology's latent culture of early belief, meritocratic discovery, and earned reputation.

a16z对谈指出，在科技生态中先于市场发现卓越人才与突破性创意的能力是最核心的竞争力；随着传统体制文凭贬值与AI带来海量申请噪音，由行业同行押注个人声誉的公开“联署”（Cosign）正成为穿透信息不对称、沉淀持久职业信誉的关键机制。 体制学历解耦与扁平社交网络的失真 (0:53) Erik Torenberg指出，传统大学教育打包了教学、社交和文凭功能，但在互联网时代，公开的工作成果证明（proof-of-work）正将其逐步解耦。作家David Perell发现其Twitter账号带来的职业价值远超花费25万美元获得的大学学位；在当下科技界，获得Patrick Collison、Elon Musk或Marc Andreessen等领袖的关注，其声誉权重往往超过斯坦福大学的文凭。这种同行联署释放出极高维度的注意力信号，能为新兴人才打破行业壁垒，也为早期发掘者积累强大的社会资本。 与此同时，现有主流职场网络难以满足精准筛选的需求。LinkedIn等平台将人际关系简化为单一、对称的双向好友关系，Josh Elman与Olivia Moore观察到，其连接已退化为泛泛之交的低信噪比数字游戏，约90%的引荐请求得到的回复仅是“我其实不太认识他们”。对于面临多家初创公司选择的初入行工程师或寻求关键高管的创始人而言，此类扁平化记录完全无法反映其真实能力，造成严重的职业信息不对称。 细粒度背书与异步意图的机制设计 (4:06) 为还原硅谷信任流通的真实机制，新一代职业信誉平台Cosign围绕三类层级化的背书原语构建：塑造职业生涯的导师关系（如Josh Elman对Erik Torenberg早期的指导）、共同经历初创公司危机的高压“战友”（In the Trenches），以及对新兴潜质人才的早期押注（Person to Watch，例如Torenberg在7年前挖掘当时年仅22岁的Tesla工程师、现Cognition总裁Russell Kaplan）。此外，平台允许从业者记录对种子期公司的早期判断，通过社会资本分配展现预测品味。 该体系进一步通过私密意图与双向确认机制解决社交摩擦。David Booth与Torenberg指出，传统引荐往往面临被拒的尴尬与认知负荷，而Cosign允许用户在后台记录具体意图（如“若其离职愿立即录用”或“若其创业愿参与投资”），唯有双方意向契合时才触发引荐。针对社交平台X上赞誉仅留存24小时便被时间线吞没的缺陷，平台通过问答与重大里程碑公告将瞬时称赞锚定为永久档案，使从业者无需依赖易衰退的人脑记忆，即可跨越时间成为规模化的人才撮合者。 幕后操盘手价值与AI时代的信任稀缺 (25:55) 硅谷历来将聚光灯不成比例地投向创始人，而大量决定公司成败的幕后操盘手却长期缺乏辨识度。Olivia Moore指出“4号工程师难题”：许多不可或缺的技术与产品骨干在离开原团队创立新项目或跳槽时，外界难以查验其真实贡献，导致个人声誉流动性受阻。Cosign的前身源自Torenberg旨在表彰非创始人工程师和设计师的RISE Awards，通过抓取全网公开履历并引入AI结构化建档，自动沉淀被同行验证的执行细节，解决网络启动的冷启动困境。 在大模型与自动化工具普及的背景下，真实人类背书的稀缺性愈发凸显。Moore分享的一项实验显示，仅在LinkedIn发布一条简单的初创企业求职表单，便瞬间吸引了2000份申请；随着自主浏览器代理等AI工具泛滥，未来单条招聘可能触发数以万计的自动投递。当申请与外联成本趋近于零、噪音呈指数级爆发时，经过时间检验的人际协作记录、高风险场景下的共事经历以及具名的人类背书，构成了穿透虚假信号的终极稀缺资产。 早期相信者的长期复利与阿尔法争议 (48:09) 针对“公开发掘顶尖人才是让渡私有阿尔法（Proprietary Alpha）”的质疑，David Booth认为该逻辑仅在极短交易周期内成立，公开联署的真正杠杆体现在未来两到五年的复利中。在市场尚未形成共识前公开押注一位新秀，能永久确立自身作为其“最早相信者”的地位。历史经验表明，早期行业巨头的公开背书具有造王者效应：Paul Graham在2009年便将未有标志性退出事件的Sam Altman与Steve Jobs并列；Ben Horowitz对Databricks学者创始人的早期力挺；以及Dwarkesh Patel在二十几岁即凭借Collison与Andreessen的公开支持使播客成为顶流。 这种早期支持具有显著的不对称回报。创始人对其最困难、最边缘时期给予信任的人抱有极高的忠诚度，其心理权重远高于后期锦上添花的大额资本。正如Brian Chesky在Airbnb上市公开信中，越过多轮后期巨额投资机构，将首要敬意与感激归于Y Combinator与Paul Graham。公开发表对后起之秀的高置信度判断，不仅能在其未来融资与招募时锁定排他性信任，还会成为向整个生态展现自身鉴赏力的磁石，持续吸引下一批卓越构建者。 尽管Cosign力图通过透明公开的信誉网络打破传统壁垒，但对谈者亦承认，由庞大且安于现状的在职从业者所支撑的二十年网络效应与平台防护壁垒，使得LinkedIn等巨头难以被直接颠覆。Cosign的核心定位并非正面替代传统通用招聘软件，而是作为聚焦于科技创新圈的信誉基础设施；其长期效用仍取决于能否在AI生成的大量信息中，持续维持具名同行背书的高准入门槛与声誉约束机制。

## Background 该视频由知名风险投资机构 a16z 于 2026 年 9 月 25 日发布在 YouTube 频道。视频中主持人 Erik Torenberg 与 Josh Elman、Olivia Moore 及 David Booth 深入对谈，正式介绍了围绕职场声誉构建的新产品 Cosign，探讨如何让行业内的隐性信任、早期背书与人才信号更加显性化与持久化。 ## TL;DR 对话深入探讨了硅谷最核心的隐形壁垒——先于市场发现卓越人才与优质项目的能力。嘉宾们指出传统凭证和 LinkedIn 的扁平化连接在 AI 泛滥时代正面临严重的信号稀释，真正稀缺的是人类带着信誉抵押的真实背书。通过介绍新产品 Cosign 的导师、战友与潜力新人三大维度及私密意向机制，团队阐述了如何打破传统履历模式，让幕后操盘手和新秀的声誉更具流动性。最后，对谈强调了“成为最初相信者”的超长期复利价值，论证了公开背书不仅不会泄露投资优势，反而是构建个人社会资本的正和飞轮。 ## Atmosphere & style 整场对谈充满着硅谷资深内行围炉夜话式的坦诚与敏锐，既有顶尖投资人对行业潜规则的精准拆解，又不乏同行老友互相调侃与共鸣的松弛感。嘉宾们的语调兼具创业者的理想主义与实干家的清醒，从当年 Rise Awards 的往事娓娓道来，在严谨的技术思考与生动的迷因段子间自然切换。整场对话散发着强烈的建设者氛围与智识碰撞的快感，让人切实感受到一股推动声誉民主化、为下一代建设者喝彩的由衷热情。 ## Bulletin - [00:53](https://www.youtube.com/watch?v=-ywZlfznTa4&t=53s) a16z 团队推出围绕职业声誉构建的新产品 Cosign，旨在将科技生态中隐性的人脉信任显性化与结构化，打造高信噪比的人才与早期公司网络目录。 - [34:25](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2065s) 在自动化脚本与 AI 带来海量低成本求职申请的背景下，人类基于真金白银与深刻认知的真实背书正在成为最极致的稀缺资产。 - [04:06](https://www.youtube.com/watch?v=-ywZlfznTa4&t=246s) 针对 LinkedIn 泛化连接导致信用稀释的痛点，Cosign 引入导师、生死战友、潜力新人三大核心维度及私密意向匹配，精准捕捉高颗粒度职业信号。 - [48:09](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2889s) 硅谷本质上由精英唯才制驱动，打破“保留独家优势”的零和思维并成为潜力人才“最初的公开相信者”，能在未来数年内沉淀出指数级放大的长期声誉复利。 ## Key takeaways - [11:05](https://www.youtube.com/watch?v=-ywZlfznTa4&t=665s) 在科技创业与投资领域，最核心的能力是先于市场发现事物与人才的敏锐前瞻力，但以往这种眼光缺乏公开持久的记录机制。 - [00:53](https://www.youtube.com/watch?v=-ywZlfznTa4&t=53s) 点对点的行业巨头关注与背书，其产生的声誉溢出效应已远超传统大学文凭等机构认证的价值。 - [04:06](https://www.youtube.com/watch?v=-ywZlfznTa4&t=246s) Cosign 通过“塑造职业生涯的导师”、“并肩作战的战友”和“最值得关注的潜力新人”三个颗粒化维度，防止普通社交网络泛化关注导致的信用稀释。 - [04:06](https://www.youtube.com/watch?v=-ywZlfznTa4&t=246s) 普通从业者可以通过公开背书早期种子轮公司进行“社交资本投资”，凭借长线眼光累积自身声誉与伯乐地位。 - [06:18](https://www.youtube.com/watch?v=-ywZlfznTa4&t=378s) 初入职场的优秀人才面对初创企业 Offer 往往缺乏辨别真伪的渠道，高质量的声誉网络能够大幅降低市场的信息不对称。 - [12:52](https://www.youtube.com/watch?v=-ywZlfznTa4&t=772s) “私密意向信号”允许用户低门槛标记“创业即投”或“离职即招”，在消除社交沟通摩擦的同时完美解决人脉撮合中的时机错位。 - [18:37](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1117s) 动态流聚焦于“重大公告”与“具体提问”，将原本在社交媒体上转瞬即逝的优质评价固化为可长期沉淀的声誉资产。 - [24:32](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1472s) 科技生态需要摆脱光环仅归于创始人的局限，通过公开背书让幕后的卓越操盘手、核心工程师与设计师能够带着真实战绩流动。 - [36:50](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2210s) LinkedIn 的致命短板在于将二元连接一视同仁，而其历经 20 年构筑的坚固壁垒主要依赖于数亿“对现状轻度满意但开放机会”的职场人黏性。 - [40:51](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2451s) Cosign 利用 AI 抓取全网公开数据预先构建档案，打破冷启动壁垒，并借鉴维基百科由他人撰写词条的声望模式赋能用户。 - [46:18](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2778s) 硅谷是允许跨越资历壁垒的唯才竞技场，年轻后辈凭借惊艳洞见获得资深前辈的公开 Co-sign 即可迅速实现圈层跃迁。 - [48:49](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2929s) 将独家发现（Alpha）私藏不如尽早公开背书；人们对在默默无闻阶段就坚定相信自己的支持者，其感恩与长期回报会呈现高出数个量级的爆发。 ## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=-ywZlfznTa4&t=0s) 硅谷最核心的技能与声誉的价值传递 对市场趋势与杰出人才的先发洞察，是科技投资与创业领域最关键的壁垒。对话伊始便直指核心：“在这个行业、甚至在整个世界上，最重要的能力就是先于市场发现事物的能力。”世界上存在着大量渴望相互结识的人才，双方之间横亘的仅仅是信息差。在硅谷，人们对公开发表背书有着极为苛刻的标准，而像 a16z 这样的顶级风投机构，其日常核心工作正是挖掘优秀人才并将他们引入生态网络，让信任与声誉随着时间的推移不断沉淀与放大。 嘉宾们借用了 Naval Ravikant 的名言——“做到足够优秀，以至于人们借由你来做生意”（"Be so good that people do business through you"）。在理想的声誉网络中，如果一位极具辨识力的人同时为两个人提供信用背书，即便这位推荐人完全没有居中牵线，这两人也能基于背后的信任基础建立联系并共同创造出新的事业。然而，在现有环境下，要在正确的时间将特定技能与需求精准匹配依然极度困难。 ```mermaid flowchart TD accTitle: 硅谷声誉与信用传递网络机制 accDescr: 展示了个人声誉如何通过公开背书形成信任传递，并促成新的合作与发现。 subgraph SignalSource["信用源头与背书"] A["行业领袖 / 资深导师"] -->|"公开背书 (Cosign)"| B["被背书者 A"] A -->|"公开背书 (Cosign)"| C["被背书者 B"] end subgraph AutonomousConnection["自发连接与价值溢出"] B <-->|"基于共同背书建立信任"| C C --> D["自主诞生合作与初创项目"] end subgraph FeedbackLoop["反向赋能"] D -.->|"验证眼光，增强社交资本"| A end ``` --- ### [00:53](https://www.youtube.com/watch?v=-ywZlfznTa4&t=53s) 推出新产品 Cosign：点对点背书与非传统信号 Erik Torenberg 正式介绍了新产品 **Cosign**。这是一款围绕创业生态专业声誉构建的工具，旨在打造关于公司与人才的最完备网络目录。Erik 以知名创作者 David Perell 的经历切入：David 曾感慨自己的 Twitter（X）账号比花费了 25 万美元的大学文凭更具价值，无论在职业发展还是个人成长上都给予了他更多助力。传统大学本质上是将“教育、人际网络、文凭认证与社交生活”打包捆绑在一起的产物；而在社交网络时代，人们通过日常探讨获取教育，通过公开输出证明自身实力（Proof of Work）并结识同行。 谈及文凭的去中心化解绑，Erik 抛出了一个思想实验：“去斯坦福大学念书的人，和被 Patrick Collison、Elon Musk 以及 Marc Andreessen 关注的人，哪一个更令人印象深刻？”绝大多数人的答案显然是后者。这种关注本质上就是一种点对点的、无需传统机构认证的显性或隐性背书（Cosign）。哪怕这些行业巨头只是想阅读某人的内容，这种关注所产生的“信誉副产品”（Reputational Exhaust）也足以打开所有大门，引来招聘、投资与合作机会。 这种信号机制具备双向受益的特征。对于被发现的人而言，这无疑是进入核心圈层的通行证；而对于背书者而言，率先挖掘出尚未成名的新秀同样是积累自身声誉的核心途径。硅谷的迷人之处在于“你永远不知道下一个改变行业的大人物会是谁”，他们可能只有 18 岁，或者来自毫无名气的高校。正因如此，像 Zach Frankel 这样因善于先发发掘顶级人才而建立声誉的投资人，便能获得人才的青睐与“造王者”的地位。在传统的风险投资与天使投资之外，仍有大量既非创始人也非投资人的人才渴望表达这种对彼此的认可，而 Cosign 正是为这种广泛的信用信号提供了载体。 --- ### [04:06](https://www.youtube.com/watch?v=-ywZlfznTa4&t=246s) Cosign 的三大核心背书维度与“社交资本投资” 为了避免泛化社交平台（如仅仅点击 Follow）所带来的信号稀释，Cosign 在每个用户个人档案中设计了三个最为核心的背书维度，力求将隐性的人脉信任显性化、结构化： 1. **谁塑造了你的职业生涯（Who shaped your career / 导师）**：硅谷从根本上依赖导师文化运转。Erik 举例提到，自己年轻时 Josh Elman 曾投入大量时间指导他，这一事实本身就是对 Erik 能力的强力信用传导——“如果你尊敬 Josh，就会因为这种联结而顺带认可我”。同时，愿意倾囊相授的导师也会因此赢得声誉，吸引更多未来之星主动寻求合作。 2. **愿在任何地方共事/并肩作战的战友（Would work with anywhere / In the trenches with）**：现有的职场社交平台极难反映出一个人真正愿意与谁“上战场”。能够明确指出自己在高压环境下与谁建立了生死与共的信任，是远比普通好友关系更加颗粒度细腻的高信噪比信号。 3. **最值得关注的潜力新人（Person to watch）**：由行业中人指认出自己认为即将崭露头角、实现爆发式成长的新生代人才，并允许用户对其进行排序。Erik 透露自己曾在七年前给数百人发送邮件征集这一问题，其中有人提名的 Russell Kaplan 当时只是特斯拉一名 22 岁的普通工程师，如今他已成为前沿 AI 独角兽 Cognition 的总裁。 除了对个人的背书，Cosign 还将这一逻辑延伸至早期公司。用户可以公开标注自己在种子轮阶段就坚信会成为行业巨头的公司。对于普通从业者而言，这相当于一种“梦幻投资（Fantasy Investing）”或社交资本层面的公开下注，让非投资人也能借由证明自己的眼光来累积声誉。 --- ### [06:18](https://www.youtube.com/watch?v=-ywZlfznTa4&t=378s) 硅谷内部的信息不对称与人才招聘痛点 Olivia Moore 结合自己近十年的风险投资经历表示，硅谷中最宝贵的数据莫过于“谁真正拥有坚定信念”（Who has conviction），以及“谁能提供关于一个人的真实评价”。当下即便顶尖的风投机构每天动用全部资源四处向人打听候选人的背调情况，依然要付出巨大的心力才能淘选出有效信号。这种巨大的信息不对称在普通求职者身上体现得尤为残酷： - **职场新人的决策困境**：一名刚入行的初级工程师面对五个初创公司的 Offer，完全没有信息渠道去判断哪家是真正的“火箭飞船”，哪家早已深陷泥潭。选对或选错，其职业生涯的发展轨迹将天差地别。 - **信息壁垒的普遍性**：Olivia 分享道，自己的双胞胎妹妹 Justine Moore 目前在 a16z 负责基础设施投资团队。两人虽从斯坦福大学毕业并曾参与运营斯坦福的创业孵化器，但在真正进入风投机构之前，依然根本无从辨别市面上众多初创公司与创始人究竟谁是真材实料。 - **长达近七年的精选实践**：在初入风险投资行业时，Olivia 与 Justine 创办了一份每周日风雨无阻推送的创业招聘周报，持续运营了长达六七年之久。正是因为两人掌握了风投机构内部的“内幕信息”，才得以精选出值得年轻人加入的高潜力初创公司，并在全美高校建立了庞大的校园球探（Scout）网络，这也从侧面印证了硅谷圈外对高质量信誉背书数据的饥渴。 ### [08:37](https://www.youtube.com/watch?v=-ywZlfznTa4&t=517s) 降噪时代下的声誉背书与活体花名册 在当下的创业环境中，创办一家公司的门槛变得前所未有地低，但与此相伴的是海量噪音的涌现。对于一名刚毕业走出校门、面对纷繁初创生态的学生而言，筛选信息的难度呈指数级上升。他们不仅需要依赖风险投资机构的公开投资背书，更需要依赖行业资深人士在投资轮次发生前就给出的早期信号。Cosign 的核心使命，正是让更多人能够规模化地将个人声誉押注在他们认同的团队和想法之上。 以知名科技创作者 Packy McCormick 为例，他在深科技（Deep Tech）和垂直整合型企业（Vertical Integrators）领域拥有极高水准的敏锐嗅觉与信号识别能力。在 Cosign 平台上，Packy 精心挑选并背书了 37 家公司，直言“如果你想进入这个领域，就去这些公司”。这种行为本质上是将他个人的声誉与这 37 家企业深度绑定——他不仅公开撰写深度分析，更用真金白银和自身名誉为之担保。这使得 Cosign 超越了传统的名片工具，演变成一份动态的、持续更新的“高信噪比公司与人才活体花名册”。 与收取高昂订阅费的商业数据库（如 Crunchbase、PitchBook 等）不同，Cosign 选择对创业生态完全免费开放。平台的定位非常纯粹：只展示积极且准确的信号，打造一个让行业顶尖建设者和投资人以极高精度展示成果的聚集地。团队明确做出了产品取舍：不求迎合所有人，但必须彻底解决传统职场平台的“信号稀释”痛点。在 LinkedIn 上，两个人互为好友往往并不代表真实关系，私信询问时往往只会得到“我其实不太熟”的尴尬回复；而在 Cosign 上，每一次背书都有极其颗粒化的真实权重。 - **高信噪比信号源**：Packy McCormick 等关键意见领袖背书的 37 家垂直整合/深科技先锋企业，为行业新人提供了无可替代的职业指引。 - **免商业化壁垒**：不同于 PitchBook 和 Crunchbase 等昂贵产品，Cosign 以免费生态服务于初创社区，透明呈现 Olivia Moore 等投资人的投资组合与企业核心团队构架。 - **产品核心取舍**：坚持只展示正面（Positive）、真实且颗粒度极高的硬核声誉，拒绝无效泛社交。 --- ### [11:05](https://www.youtube.com/watch?v=-ywZlfznTa4&t=665s) 先于市场发现事物的能力与声誉资产化 David Booth 提出了一个贯穿硅谷乃至商业世界的核心论点：全行业最关键的一项核心技能，就是“先于市场发现事物价值的能力”（identify things before the market）。对于风险投资人而言，这项技能体现为早期发掘伟大的公司；对于创始人或招聘主管，体现为挖掘潜在的顶级人才；对于企业采购负责人，体现为识别出更具潜力的工具与产品。然而在现有世界里，这种敏锐的前瞻眼光缺乏永久的记录机制——并没有一个“小众时髦人士排行榜”（hipster leaderboard）来公开记录是谁最先发现了那家日后爆火的独立咖啡馆。 在风险投资领域，天使投资人追求“非共识且正确”（non-consensus and right），并在 Pre-seed 轮次获得高额财务回报。但长期以来，个人在社交层面的“社会资本”（Social Capital）却无法以类似的方式沉淀并复利增长。当一位行内资深人士极度信任某个默默无闻的新人时，他唯一能做的就是凭借自身信誉向 Josh Elman 这样的顶级投资人做引荐（Warm Intro）。投资人愿意抽出时间回复陌生人的冷邮件（Cold DM），纯粹是因为背后有 David 等可信背书人在拿声誉做担保。 Cosign 试图将这种隐性的社会资本显性化、机制化。当一个人获得资深人士的公开 Cosign 时，这份信用凭证就成了硬通货，甚至能直接打通后续的协作网络。David 引用了一句深得他心的经典格言：“要做到足够优秀，让别人愿意通过你来做生意”（Be so good that people would do business through you）。当平台上两个人因为同一位关键节点的信任背书而相互认识、共同创业或相互投资时，社会的信任轮盘便被彻底润滑，甚至在最初引荐者不知情的情况下持续自发运转。 ```mermaid flowchart LR accTitle: 早期发现与声誉资本流转机制 accDescr: 展示敏锐洞察力如何转化为平台公开 Cosign，进而无缝促成信任传递与商业合作的流程。 subgraph Discovery ["早期识别能力"] D1["发现被低估的人才 / 公司"] --> D2["先于市场（非共识且正确）"] end subgraph CosignToken ["社会资本凭证化"] D2 --> C1["在 Cosign 平台公开背书"] C1 --> C2["背书人声誉资产沉淀"] end subgraph NetworkAction ["信任传导与协作"] C2 --> N1["降低冷启动信任摩擦"] N1 --> N2["跳过引荐直连顶级资源"] N2 --> N3["生态自主衍生合作与投资"] end ``` --- ### [12:52](https://www.youtube.com/watch?v=-ywZlfznTa4&t=772s) 降低匹配摩擦：双向意向与颗粒化职业信号 对话随后深入探讨了 Cosign 的核心功能设计逻辑。Erik 回忆起 2018 年参加 SV Angel 年会时的难忘经历：主办方在活动前向所有受邀嘉宾发送了一份包含 150 名核心行业精英的名单，参会者只需右滑（Swipe）选择自己渴望结识的人；一旦双方互相右滑，系统就会自动完成双向引荐并附带完整背景上下文。Erik 当时几乎右滑了所有人，并在现场迅速与数十人建立了深度对话，他们日后均成为了挚友或紧密合作伙伴。这一经历证明，世界充满着“我想见你，而你也恰好想见我”的潜在契机，阻碍连接的仅仅是信息不对称与社交心理门槛。 在现有的社交机制下，主动发送一条私信往往伴随着巨大的认知负荷：必须绞尽脑汁编排话术，担心遭到拒绝，还可能显得不够得体。Cosign 在此基础上更进一步，推出了“私密意向信号”（Privately Signal）机制，将职业维度的诉求拆解为超高颗粒度的条件： - **创业即投**：“如果他们未来创办公司，我愿意出资投资”（I would fund them if they start a company） - **离职即招**：“如果他们考虑离职，我渴望招募他们”（I would hire them if they consider leaving） - **定向会面**：“在特定议题上极度渴望与对方交流” 这种被动的双向意向匹配彻底消除了主动沟通时的被拒风险与社交摩擦。对于招聘管理者而言，他们心中通常有一份极度想要招揽的人才清单，但碍于对方当前正在任职，直接私聊既耗费时间又显唐突；而通过 Cosign 的高意图被动匹配，只要对方也释放了流动信号，引荐便会在毫无尴尬的前提下水到渠成。 --- ### [15:27](https://www.youtube.com/watch?v=-ywZlfznTa4&t=927s) 跨越时间的持久记忆：重构硅谷的信任基础设施 Olivia Moore 强调，Cosign 的诞生让每个人都能转型为高效的“人才连接者”（People Matcher）。在日常工作交流中，每个人都会频繁遇到能力极强、具备独特专长的顶尖人才；然而三到五个月之后，当有其他创始人提出此类专业需求时，人们往往难以瞬间在大脑中调取回忆，不得不反复确认对方是否还对外开放沟通。传统的“脑力记忆匹配”极其脆弱且受限于短期热度，而 Cosign 则扮演了一个跨越时间的持久化结构性数据库。 Josh Elman 对此深表赞同。硅谷的核心文化由“促成连接”构成，但最大的瓶颈始终是“在极其精准的时间节点，将供给方的技能与需求方的痛点完全对齐”。以往，当有人向 Josh 求助“你能否帮我找一个擅长应对公关危机的 PR 专家，或指导我做职业抉择的导师”时，Josh 只能在最近聊过的人脉圈里搜寻。而现在，求助者可以直接翻阅 Josh 的 Cosign 记录，看到某位专家曾在哪次危机中提供过至关重要的协助，再向 Josh 提出确认：“我看到你曾背书过这位专家在这件事上的表现，你觉得他愿意帮我吗？” 这种机制在不稀释社交圈信任厚度的前提下，实现了声誉的复利与规模化传递。它既不是完全不受控的开放式信息广场，也不是封闭不通的私人通讯录，而是一个让信任可以安全流动、精准沉淀的声誉基础设施。 ### [16:38](https://www.youtube.com/watch?v=-ywZlfznTa4&t=998s) 私密与公开双轨信号：解决人脉网络中的时机错位 David Booth 指出，Cosign 在设计上融合了“私密信号”（private signal）与“公开信号”（public signal）。整个产品的探索过程充满趣味，就像在浏览维基百科或使用 Pinterest 一样，用户可以在人物、个人资料、公司以及投资人之间不断顺藤摸瓜。在这个过程中，用户可以建立双重清单：一方面是将心仪的人才放入个人的私密清单，标记为“未来想要共事的人”；另一方面则是整理出公开列表，例如自己极度认可的优秀设计师清单。这样一来，每当有创业者前来求助询问“我需要一个新设计师、增长操盘手或工程师”时，推荐者既能直接调取私密储备，也能大大方方分享公开清单。这不仅向被推荐者及其网络传达了坚定的信任，更为求贤若渴的创始人提供了捷径，免去了每次重复推荐的沟通成本。 Olivia Moore 随之指出了职业网络中长期存在的一个痛点——“时机错位”（timing discrepancy）。在现实中，投资人或朋友可能敏锐地察觉到某家初创公司举步维艰，而另一家公司正渴望收购他们或招揽其团队。然而，若直接跑去询问创始人“你是否考虑出售公司”，往往蕴含极高的社交风险，甚至可能被视作冒犯；毕竟对方的公司可能仍在艰难运转，创始人也希望继续拼搏。Cosign 这种双向网络机制让双方都能自主沉淀并表明意向，在时机成熟时自然撮合人才招聘、公司并购或职业转型。对话中大家还打趣道，连“公司买卖”这种突发奇想都快要被加进明天的产品功能路线图里了。 ```mermaid flowchart TD accTitle: 双轨信号与动态撮合架构 accDescr: 展示私密与公开清单如何打通人才、创始人与机遇之间的双向流转 subgraph UserList["用户背书库"] Private["私密清单<br/>（未来共事意向 / 潜在并购线索）"] Public["公开清单<br/>（精选设计师 / 工程师 / 增长专家）"] end subgraph Triggers["触发场景"] Ask["创始人寻人求助<br/>（急需核心岗位）"] Timing["契机出现<br/>（转型 / 融资 / 出售意愿）"] end subgraph Outcomes["撮合结果"] Private --> Timing Public --> Ask Timing --> Match["精准无损对接<br/>（求职、并购、合作）"] Ask --> Match end ``` ### [18:37](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1117s) 动态流的双核驱动：让碎片化的社交赞誉沉淀为持久声誉 在讨论产品信息流（Feed）的构建时，团队曾反复权衡是采用完全自由的格式还是加以约束。在 Josh Elman 的建议下，Cosign 最终将动态流聚焦于两件事：公告（Announcements）与提问（Questions）。现存的职业声誉大量散落在社交媒体上，但它们极其转瞬即逝（ephemeral）。比如 Josh Elman 最初加入 a16z 时，许多曾与他共事多年的伙伴纷纷现身公开称赞：“Josh 曾经在我的职业生涯中给予巨大帮助，你一定要和他合作。”Erik Torenberg、Nikita Bier 以及众多业内人士都发出了由衷的肯定。然而在传统社交网络上，这类高价值背书通常在第二天就被信息洪流彻底淹没。Cosign 的目标就是将这些高光时刻固化在个人主页上，使其转化为经久耐用的“声誉资产”（durable reputation）。 在此框架下，每次公司融资或人才跳槽，都是发起背书与 Co-sign 的黄金契机；而提问板块则聚焦于极其具体的诉求，例如“谁是处理此类危机的最佳公关顾问”或“金融科技领域最顶尖的天使投资人是谁”。当某人因回答而被列入推荐清单，其专业声誉便得到了一次可沉淀的加固。Erik 形象地引用了那个经典的火柴人派对迷因（“他们甚至不知道我……”）打趣说：当你获得了一位极具分量的背书人——比如马克·安德森（Marc Andreessen）为你入职撰写了官方博客——那种自豪感是无与伦比的，任何人都巴不得让全天下第一眼看到自己时就能感知到这种顶级信任。让这些散落的节点变得清晰易读（legible），正是还原一个人真实专业分量的关键。 - **动态流的两大核心支柱**：聚焦于“重大公告”（融资、跳槽等里程碑）与“具体提问”（寻找特定领域最优秀的人选）。 - **对抗瞬时遗忘**：利用机制将 X 等平台上原本会迅速沉没的高质量评价捕获并固化在主页。 - **声誉资产化**：背书不再是一次性的社交客套，而是伴随职业生涯终身可查的信任凭证。 ### [21:38](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1298s) 庆祝文化、信誉抵押与照亮幕后功臣 当团队接入互联网的实时数据流后，才真正意识到行业中每天都在发生海量的融资官宣和人才流动。Cosign 团队通过筛选 X（前 Twitter）上的引用推文与回复，提炼出真正真诚、有深度的正面评价，将其精准链接至评价者与被评价者的个人主页。团队强调，Cosign 坚持主打正向维度的“庆祝文化”，因为产品想要胜出就必须在特定维度做到极致，而 Cosign 旨在成为人们展示自我与团队最高水准的舞台。打造一款产品往往凝聚着无数心血、爱意、熬夜写下的代码（无论是由人还是由 AI Agent 协助完成），将其公之于众本身就是极其脆弱的瞬间，这种付出理应被长久铭记与喝彩，而不是让其湮没在四处可见的苛刻批评中。 更深层次的逻辑在于，硅谷精英圈层对“公开背书”有着极高的心理门槛与信誉抵押机制。公开推荐一个不够优秀的人，会直接反噬推荐者自身的声誉，因此人们对转介绍和站台极为慎重。这种自发的声誉约束机制保证了背书的含金量，用户很快就能分辨谁是在不负责任地盲目吹捧，谁是真正体验过产品、深知背后艰辛后给出的硬核认可。此外，对话特别强调了打破“创始人独占光环”的必要性。创立一家公司固然伟大，但要把愿景变为现实，离不开整个团队的合力支撑——无数在幕后默默奉献的卓越设计师、工程师、产品经理、市场人员与财务专家，同样值得站在聚光灯下，获得与其贡献相匹配的持久声誉认可。 ### [24:32](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1472s) 幕后功臣与 Rise Awards：为“隐形贡献者”正名 一家伟大公司的构建包含无数细微的环节，但长久以来，聚光灯几乎全打在创始人和高管身上。大家一致认为，针对具体团队或关键个人的背书与联署（Co-sign），其力量将远远超出行业日常关注的名人效应。David 借此抛出了一个十多年前的渊源，促请 Erik 聊聊早期的 Rise Awards。大约在 2015 至 2016 年间，正是这个奖项让大家初次结识 Erik。当时的核心论点就是：创始人拿走了所有光环与赞誉，但真正把事情做成的，往往是那些藏在幕后的操盘手（Operators）。 Erik 回忆道，Rise Awards 是专为工程师、设计师和产品经理设立的奖项，Josh 当时也是评委之一。该活动收到了数千份申请，挖掘出了许多后来大放异彩的顶尖人才，而他们如今正打算以某种形式重启这个项目。Erik 引用了李小龙的名言打趣道：“不怕练过一万种踢法的人，只怕把一种踢法练了一万次的人。”正因如此，Rise Awards 和 Cosign 的底层逻辑在他们的脑海中已酝酿了十余年。更重要的是，正如 Olivia 指出的，当年获奖的人通过闭门退修会（retreats）和专属社群彼此结识，缔结了持续终身的深厚友谊。 这种人脉连接揭示了硅谷的一个普遍现象：很多人从明星公司离开去创业，当你深入做背景调查（Reference Call）时才会惊觉，某位当年的“4号工程师”实际上是这家公司能否成功的决定性力量。但正如之前讨论的“可识别性”（Legibility）问题，这些战绩在过去是无法被公开量化的，他们不可能成为每篇新闻报道的封面人物。如果这些关键人才离开公司去创业或加入新团队，他们理应能够完整带着这些勋章和成就同行，而现有的职业网络根本无法满足这一点。 --- ### [26:36](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1596s) 让声誉更高效、更具流动性：消除人才市场的信息不对称 针对这种痛点，Cosign 旨在让“声誉市场”（Reputation Market）变得更加高效且具有流动性。Erik 分析称，目前市场上存在巨大的信息不对称：有些人拥有极其过硬的行业声誉，但仅限于私下极窄的圈子，外界对其缺乏认知；如果这份过硬的声誉能被透明化、流动化，就会有源源不断的资本想投资他们、顶级企业想雇佣他们。 在实际操作中，这种流动性还能解决求职与招聘的双向撮合痛点。很多顶尖人才在考虑换工作时，经常会私下询问投资人：“嘿，我到底该去哪家公司？”如果投资人能直接给出自己公开盖戳认证（stamped）的推荐公司名单，求职者便能放心对接；甚至只需一键操作，人才的信息就能定向推送给投资人认可的所有公司。反过来，当一家公司急需招聘公关主管（Head of Comms）时，他们无需大海捞针，只需看圈子里有哪些信任的人公开 Cosign 过该领域的佼佼者，整个行业的人才匹配效率将获得指数级提升。 投资人自身也有强烈的表达需求。Olivia 坦言，目前投资人对自己所投公司的背书极为低效，大多只能挤在 X（原 Twitter）的个人简介或 LinkedIn 上，或者被埋在风投机构官网的投资组合页面里。为了不破坏自己在社交平台上的粉丝粘性和受众体验，投资人不可能每天疯狂发帖给每一家被投公司打广告，尽管他们极度渴望能有一个专门的阵地，能够“毫无顾忌、光明正大地”全天候为自己坚信的团队站台背书。 ```mermaid flowchart LR accTitle: Reputation Liquidity Flow accDescr: Diagram illustrating how invisible operator value is converted into liquid reputation through Cosign. subgraph Invisible["传统模式：信息孤岛"] A["幕后核心贡献者<br/>(如 4号工程师)"] -->|深埋幕后| B["仅限少数同事知晓"] B -->|缺乏流通手段| C["换工作/创业时<br/>信誉无法完整迁移"] end subgraph Liquid["Cosign 模式：声誉流动化"] D["同行/投资人公开联署<br/>(Co-sign 背书)"] --> E["高信噪比声誉图谱"] E --> F["高效撮合：资本/招聘定向流动"] end Invisible -.->|解决不对称| Liquid ``` --- ### [28:30](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1710s) 现有平台的痛点：X 的低信噪比与 LinkedIn 的爱恨交加 在探讨如何构建新机制时，大家不可避免地审视了当下的两大巨头平台：X 和 LinkedIn。X 固然是一个伟大的平台，其溢出价值（exhaust）极高，但它根本不是为了精准发现人才或建立信任信号而设计的。Erik 吐槽道，有时候自己发布了一条招聘动态，获得了几千个点赞，但他却根本无法方便地过滤或下载这些关注者数据：在这些点赞和关注的人里，究竟谁是想来全职工作？谁是想寻求投资？谁只是顺手划过？平台缺乏清晰的意图信号。 与此同时，LinkedIn 虽然是一项极其庞大的基础设施，大家每天都在使用，但用户体验却饱受诟病，许多人对其充满抱怨。尽管坊间经常调侃平台上的动态充斥着令人尴尬的吹捧（cringe feed），甚至早在信息流诞生之前，用户就对其单调的机制感到压抑。然而，LinkedIn 依然是一个标杆级产品——正如马克·安德森（Marc Andreessen）当年为其辩护，即便微软以 260 亿美元收购它时，许多人还在感叹它是不是卖得太便宜了。既然它是无可替代的职场公用事业（utility），为什么人们依然觉得它缺少了极其关键的某些东西？ --- ### [29:41](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1781s) 早期 LinkedIn 的构想：从信任图谱到“社交简历”的演变 曾于 2004 至 2005 年在 LinkedIn 早期团队核心工作的 Josh Elman 现身说法，复盘了创始人里德·霍夫曼（Reed Hoffman）当年的战略三部曲——增长、使用、变现（Growth, Usage, Revenue，简称 GUR）： ```mermaid flowchart TD accTitle: Reed Hoffman GUR Framework accDescr: The three evolutionary stages of LinkedIn from network growth to monetized products. stage1["第一阶段：增长 (Growth)<br/>搭建职场底层网络，允许用户只注册占位"] stage2["第二阶段：使用 (Usage)<br/>构建网络之上的服务，如 LinkedIn Jobs、PR 人才搜索"] stage3["第三阶段：变现 (Revenue)<br/>将网络与应用场景商业化变现"] stage1 --> stage2 --> stage3 ``` - **初始设计的高信任度网络**：早期 LinkedIn 只有单一维度的连接信号（我们是否 Connect），但其美妙之处在于背后的“信任图谱”（Trusted Graph）。当时建立连接意味着“我真正了解你，并愿意为你的人品和能力背书”。如果 David 想找 Eric，而 Olivia 认识 Josh，Josh 认识 Eric，整个链条会像击鼓传花一样，基于真实信任逐级传递引荐。 - **关系异化与信号稀释**：随着平台盲目追求网络规模，连接的门槛急剧退化。用户逐渐变成了“集邮式社交”（collecting network）——“我们在派对上见过一面”、“我们击过一次掌”、“我们互发过一封邮件”就能成为一度人脉。当好友列表变成名片收集册时，平台就彻底丧失了辨别真实信任信号的能力。 - **“社交简历”的未竟探索**：Josh 参与搭建第一代 LinkedIn Jobs 时，里德·霍夫曼一直在拷问团队：“什么是社交简历（Social Resume）？”他们希望彻底打破“求职者提交死板简历、HR 查看枯燥职位列表”的扁平模式。理想中的社交简历不仅展示申请人是谁，更会自动标注社交上下文：你公司内部有谁认识他？谁曾经和他共事过？公司里谁的密友和他有过交集？这种被真实人脉网络所注释（annotated）的求职机制，曾是改造传统招聘的重大尝试，也构成了今天重新思考声誉产品时的宝贵遗产。 ### [32:28](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1948s) 从匿名投递到“并肩作战”：职业连接的粒度缺失 当年的招聘网站大多只是一个冷冰冰的界面，求职者把简历“扔过高墙”（throwing your resume over the wall），完全不知道对面的招聘官是谁。然而，颠覆性的转折在于把招聘经理的真实个人主页直接附在岗位详情上。求职者赫然发现自己不再是面对虚无的机器，而是将要为一个具体的人工作；与此同时，系统还能展示求职者与招聘经理乃至该公司的其他员工之间存在怎样的连带关系。这种将社交图谱引入求职场景的举措，启发了人们去思考如何真正重塑职业社交网络。 但在如今的生态中，大量的真实互动仍然只短暂且碎片化地停留在 X（前 Twitter）上，而无法沉淀并注入到 LinkedIn 这种产品中。职场中最珍贵的连接往往源自“共同在战壕里打过仗”（in the trenches together）、一起合作过某个棘手项目，或是在生死存亡的危机时刻有人伸出援手拉了自己一把。当这些具备高浓度上下文的信息能够注入网络，平台的作用才会真正被激活。嘉宾们认为 LinkedIn 固然是一个伟大的偶像级产品，每天创造巨额营收，但其致命短板在于将所有的“连接”（Connection）一视同仁。LinkedIn 早期推出的技能背书（Endorsements）功能早已名存实亡；现实中，90% 的情况下如果根据 LinkedIn 主页去请求对方引荐某个好友，收到的回复往往是：“唔，我好像不太熟，可能只是在哪个派对上见过一面。”平台完全抹平了深浅关系之间的维度： - 你们是真的认识，还是仅仅萍水相逢？ - 你们是否在同一家公司工作过？工作时间是否重合？ - 你们是在同一个敏捷小组、紧密协同的相邻团队，还是纯粹的上下级汇报关系？ Cosign 的核心切入点正是将这些被扁平化为 0 和 1（认识/不认识）的二元关系，拆解回其本应拥有的高颗粒度信息层中。 --- ### [34:25](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2065s) AI 时代的投递洪流：人类背书成为终极稀缺资产 当前正处于通过技术沉淀个体隐性上下文（“社交履历 / Social CV”）的黄金时期。a16z 无论是一线的投资团队还是投后赋能运营团队，核心工作都是发掘卓越的人才、向他们投资，并将其输送对接至被投企业网络。过往的做法主要依赖从现存职场平台抓取简历，再辅以大量人工背调去摸索其背后的强弱人脉关系。然而，这种依靠外部二元信号的传统模式在当下已经举步维艰，甚至正在彻底失效。 Olivia Moore 举了一个切身案例：她最近在 LinkedIn 上发了一条动态，表示愿意帮被投公司招募早期团队成员，如果有人想加入初创公司可以填写一个 Google 表单。结果该帖子反响极其热烈，但这反而演变成了一场筛选灾难——她收到了整整 2,000 份申请。随着 AI 浏览工具和自动化申请脚本的普及，这 2,000 份很快会在一个月内膨胀到 10,000 份，乃至 100,000 份。面对如此庞大的垃圾申请洪流，即便使用像 Astra 这样的高级自动化评估工具拉满算力，HR 和投资人也根本无法梳理甄别。在 AI 使得冷启动外联和申请成本归零的时代，最极致的稀缺品变成了“人类真金白银的背书、人类的由衷信服与深度了解”。 ```mermaid flowchart TD accTitle: AI Outbound Surge vs Human Endorsement accDescr: Diagram illustrating how AI-generated applications create noise and elevate the value of human conviction on Cosign. subgraph Problem["传统渠道的痛点"] A["自动化脚本 / 浏览器 Agent 泛滥"] --> B["低成本、无限制的海量外联申请<br/>(2,000 ➔ 10,000 ➔ 100,000 份)"] B --> C["筛选系统过载瘫痪<br/>信噪比极低，二元连接失效"] end subgraph Solution["Cosign 机制与解法"] D["高颗粒度关系网络<br/>(共同经历、背书打底)"] --> E["人类真实信念与人际背书<br/>(Human Endorsement / Conviction)"] E --> F["具备极高置信度的高价值信号密度"] end C -.->|倒逼需求转移| Solution ``` 当前的社交工具往往走向两极化：一端是面向全世界漫无目的广播的公域大广场，另一端则是仅有 50 人的私密群聊，中间严重缺失一个容纳约 10,000 人、经过深度筛选互信的中间层网络。Cosign 正在填补这块空白，让每一次发帖招募或寻求合作时，系统都能沉淀出具备高度信任感的反馈。 --- ### [36:50](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2210s) LinkedIn 的二十年壁垒：轻度满意者与网络效应的护城河 讨论回到了职业社交产品形态本身。嘉宾们调侃称，人们希望职场网络能有一个像当年“MySpace Top 8”（置顶展示最好的 8 位密友）一样的产品特性：清晰标明谁是你在战壕里并肩搏杀过的战友、你日常与谁沟通、大家真正的共同好友是谁、你们如何结识、对方到底如何评价你的能力。LinkedIn 记录的是一个人的“职场身份”（Professional Identity），是名义上“你认识的人”；X（Twitter）展现的是“对你感兴趣、想认识你的人”；而 Cosign 的愿景则是构建一个关于“职场声誉”（Professional Reputation）的网络，在颗粒度上兼顾两者。 回顾行业历史，自 LinkedIn 成立以来的 20 多年里，没有任何竞品能在体量上对其造成实质性威胁。知名投资人 Charles Hudson 早在 12 年前就曾撰文呼吁反垄断或颠覆 LinkedIn，痛陈其产品体验的种种缺陷；过去十几年里，各大风投机构也投出了无数主打“LinkedIn 颠覆者”的项目，甚至有人试图用加密货币（Crypto）的代币经济学激励来瓦解其网络，但全都折戟沉沙。大家经常开玩笑说“所有人都讨厌 LinkedIn、其净推荐值（NPS）极低”，然而人们却依然沉迷且不得不频繁使用它。 Josh Elman 对此提出了不同视角：实际上，对于那些不处于频繁求职跳槽期的普通用户而言，LinkedIn 的 NPS 相当高。LinkedIn 服务的本质上是两类受众： 1. **主动找人者**：如销售、商务拓展（BD）以及风险投资人（VC），他们是平台的高频甚至重度使用者。 2. **在未来某个时刻希望被好机会找到的人**：即所谓“轻度满意当前工作者”（mildly happily employed）。 这部分“轻度满意”的人群占了压倒性多数。作为一名普通工程师，他不需要每天刷新求职动态，但他依然乐于把个人主页留在平台上。因为一旦有量身定做的高薪职位或极具吸引力的新冒险主动找上门来，他会乐于评估考量。能够将全球数亿（如今已超 10 亿）安于现状但开放机会的职场人维系在同一个网络之中，这构成了人类商业史上黏性最强、最坚固的 20 年网络效应。 ```mermaid flowchart LR accTitle: Dual Audience of LinkedIn Network Effect accDescr: Flowchart contrasting the two distinct user bases underpinning LinkedIn's sticky network effect. subgraph HighFrequency["高频找人方"] HF1["销售人员 (Sales)"] HF2["商务拓展 (BD)"] HF3["风险投资人 (VC) / 招聘官"] end subgraph PassivePool["低频池：轻度满意受众"] P1["庞大的在职专业人才库<br/>(工程师、产品经理等)"] P2["现状满意，无迫切求职意愿"] P3["保持个人履历公开以备顶级机遇出现"] end HighFrequency -->|精准搜寻与触达机会| P1 PassivePool -.->|沉淀超 10 亿用户黏性| HighFrequency ``` - **高迁移成本与防爬虫机制**：LinkedIn 的转移成本极其高昂。一方面，其反爬虫与机器人检测技术极负盛名，外部应用很难低成本实时抓取其数据（市场上甚至诞生了专门倒卖 LinkedIn 爬虫数据的成熟黑灰产生态）；另一方面，职场发展属于个人的重大人生决策，只要“其他所有人都在 LinkedIn 上”，用户就绝不可能轻易注销离开。 - **垂直生态的切入机会**：要想打破这一垄断局面，唯有通过划时代的新范式创造出全新的信息维度（例如极具穿透力的高质量人类背书），或者针对极端细分的垂直人群量身打造。例如，GitHub 在某种意义上已经是比 LinkedIn 更好、更纯粹的程序员专属“LinkedIn”；而对于渴望建立个人声誉的人才伯乐（Talent Scout）与寻求破圈的早期创业者，一个更注重真实声誉的新平台正是历史性的破局点。 ### [40:51](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2451s) 硅谷的“赠人玫瑰”文化与非零和竞争生态 Josh Elman 深入阐述了让他对 Cosign 感到兴奋的核心原因。在他看来，Cosign 根本不需要成为一个“更好的 LinkedIn”，它的价值在于将科技行业中早已暗中存在（latently happening）的信号汇聚一处。硅谷和整个科技界最独特、最宝贵的地方，就在于那种“赠人玫瑰，手有余香”（pay it forward）的信念——哪怕一个人此前并没有显赫的名声或亮眼的履历，只要他拥有独特的洞见或出色的想法，单凭这一点就能迅速凝聚一群优秀的人才和巨大的势能，从而改变世界。相比之下，LinkedIn 如今体量过于庞大，服务于各行各业形形色色的人群；科技行业完全需要一个属于自己圈子、能将瞬时发生的口碑与严肃专业深度结合的专属平台。即使 Cosign 最终只服务好科技行业本身，这也已经是巨大的成功。 Erik Torenberg 随之提到了 Benchmark 合伙人 Eric Vishria 在 Patrick O'Shaughnessy 的播客节目中的观点：当被问及对各个创投赛道的看法时，Vishria 直言“一切都会成功（It's all gonna work）”，即每个大类指数都会极其庞大。这一非零和思维同样适用于内容与网络产品。就像人们常抱怨播客已经饱和、世界不需要新播客一样，优质的新节目一旦涌现依然能大获成功，且并不会消灭其他节目；Erik 创办的“School”也不会导致斯坦福大学办不下去。同理，LinkedIn 依然会继续稳步发展，而 Cosign 则是从十年前的 AngelList 以及维基百科中汲取了深刻灵感。 - **AngelList 留下的生态空白**：十年前的 AngelList 掌控了整个初创生态的“投资人图谱”（investor graph），高度聚焦于公司、投资人和人才三者的连接。随着后来 AngelList 战略转型并逐步淡出早期的免费公开社交网络，硅谷始终缺少一个让创始人可以提前半年摸清天使投资人版图的公共基础设施（public utility）。 - **维基百科模式的灵感**：维基百科词条极具声望和吸引力，且页面内容都是他人为当事人编写的。Cosign 同样采取“为用户预先生成档案”的策略，以一种体面、带有背书性质的方式替用户自我宣传（brag on their behalf），用户如果不喜欢随时可以选择下架。 ```mermaid flowchart TD accTitle: Cosign Product Philosophy & Influences accDescr: Diagram illustrating the three conceptual pillars of Cosign: AngelList, Wikipedia, and AI Cold-Start mechanics. subgraph INSPIRATION["精神与模式灵感来源"] AL["十年前的 AngelList<br/>早期专注初创公司/天使/人才图谱"] WK["维基百科模式<br/>具声望性 + 第三方撰写词条"] SV["硅谷独有文化<br/>Pay-it-forward & 相信潜力"] end subgraph ENGINE["技术冷启动引擎"] AI["AI 自动化聚合<br/>全网抓取公开融资、推文与背书信息"] end subgraph COSIGN["Cosign 核心价值"] PROF["专属高信噪比档案<br/>他人赞誉 + 投资/合作意向"] NET["科技行业专属声誉图谱"] end AL --> COSIGN WK --> PROF SV --> NET AI --> PROF PROF --> NET ``` --- ### [44:16](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2656s) AI 打破冷启动与 Nikita Bier 式的增长杠杆 针对新平台的冷启动瓶颈，Erik Torenberg 指出，如果没有 AI，Cosign 只能走传统路线，苦苦祈求用户自行注册并手动填补履历，面临极高的冷启动壁垒。现在，团队通过 AI 技术全网检索公开发布的信息（包括融资官宣、社交媒体动态等高质量分散数据），直接用“暴力穷举”（brute forcing）的方式构建起精准详实的人才图谱。 Olivia Moore 犀利地指出了这种模式在消费级产品增长上的威力：获客最立竿见影的手段，就是通知某人“其他人正在平台上讨论你”，这正是 Nikita Bier（tbh、Gas 创始人）百试不爽的杀手锏——“你的朋友刚刚回答了一个关于你的问题”。这种正向反馈的获客机制在庞大的硅谷创投圈以前因信息分散而难以规模化执行，但有了 AI 的聚合与结构化能力后，终于成为可能。Erik 还笑着打趣称，今天刚有人在群里戏称他是“创始人社区里的 Nikita Bier”。 David Booth 强调，Cosign 收集的所有数据全部来自互联网公开发表的信息，并非黑盒操作。团队现在的态度非常开放，并不认为发布出来的产品是完美终局，而是将其视为一个公共邀请： - 邀请社区中的所有人来认领、更新或纠正自己与他人的 Profile。 - 鼓励用户在平台上 Co-sign（联名背书）自己认可的同行。 - 用户甚至可以公开 Co-sign 给团队提的产品改进建议，共同将这个公用工具打磨完善。 --- ### [46:18](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2778s) 联名背书打破资历壁垒：硅谷为何是真正的精英唯才制 Erik Torenberg 进一步将讨论升华到整个科技行业的底层运转机制：硅谷本质上是由“Co-signs”驱动的。在传统的层级社会或保守行业中，一个人往往需要按部就班熬上二十年资历才会获得话语权，但硅谷却是一个彻底的“唯才主义”（meritocracy）竞技场，年轻后辈凭借惊艳的才华和前瞻洞见，只要获得行业老牌前辈的公开 Co-sign，就能迅速实现阶层跨越。 对话中列举了大量科技史上的经典 Co-sign 案例： - **Paul Graham 与 Sam Altman**：2009 年前后，在 Sam Altman 还没做出超级巨头或实现大规模退出的早期阶段，Paul Graham 就公开发表言论，将 Sam Altman 与 Steve Jobs 并列列入全行业最优秀的五位创业者之列。这份重量级的 Co-sign 奠定了 Altman 的行业声誉，直接助力其接管 Y Combinator，并最终孵化创立了全球 AI 巨头 OpenAI。 - **Ben Horowitz 与 Ali Ghodsi**：Ben Horowitz 在早年大力公开背书 Databricks 创始人 Ali Ghodsi，帮助其从学术界走向科技商业舞台中央。 - **Ryan Hoover 与 Erik Torenberg**：Josh Elman 补充指出，Product Hunt 创始人 Ryan Hoover 早期对年轻的 Erik Torenberg 的大力信任与背书，将他引入社区领导者的位置，成为 Erik 个人事业轨迹上的重要转折点。 - **新一代年轻旗手**：Erik 指出，当下最顶尖的思想传播者之一 Dwarkesh Patel 年仅 25 岁，几乎没有上过常规班，仅凭出色的深度播客就获得了 Mark Zuckerberg 等大佬的连续做客与公开肯定；AI 赛道举足轻重的 Michael Truell（Cursor 联合创始人）也是二十多岁的青年；著名 AI 研究学者 Leopold Aschenbrenner 同样只有 24、25 岁左右。只要展现出顶尖才华，就会有人发现你、为你 Co-sign，并彻底改写你的职业生涯。 --- ### [48:09](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2889s) 破解“保留 Alpha”的利己质疑：长期主义的正和博弈 在对谈结尾，David Booth 提出了一个关于 Co-sign 机制必然会遭遇的现实质疑：**“既然这个人是我的‘Alpha’（独家超额优势），我知道他极具潜力，我为什么要把他公之于众？私藏起来岂不是对我最有利？”** 针对这种零和博弈思维，团队从长期主义的角度进行了解题： - **当期交易与跨期资本的区别**：如果某位创始人正在融当前的 pre-seed 轮，且投资人希望独占该投资额度，当下确实可以先不 Co-sign。但真正有长远价值的，是为那些在未来 2 到 5 年内才会创办伟大公司的人背书。 - **声誉复利与互惠回报**：当你在一个新人寂寂无名时公开给予 Co-sign，你实际上成为了他早期信用的支柱。当几年后他凭借这份信任找到了核心合伙人、拉齐了前五名早期员工并成功融资时，他心中永远会记得这份提携之恩——“我今天能拿到这笔融资，正是因为两年前 Olivia 为我做了 Co-sign”。公开背书不仅不会稀释认知优势，反而将私有的信息差转化为了长期稳固的社会资本与无限放大的正和飞轮。 ### [48:49](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2929s) 成为“最初的相信者”：长期复利与公开认可的巨大价值 与会嘉宾们深入探讨了以极其长期的视角投资人脉与声誉的逻辑。如果你能成为某个人或某款产品最初的相信者（early believer），并在他们默默无闻时给予支持，这种信任将在未来带来难以估量的长期回报。创业者功成名就后，往往最先联系、最先回馈的就是当初在他起步阶段就坚定支持他的人。正如嘉宾们所指出的，一旦你转变思维、抱持这种超长期主义的心态，累积的收益将呈现出数量级的爆发，因为人们永远会以高出数个量级的感激之情，铭记那些“最早相信他们的人”。 嘉宾以 Airbnb 创始人布莱恩·切斯基（Brian Chesky）为例：在 Airbnb 最终上市时，尽管在漫长的创业征程中引入了海量的后期资金、经历了无数投资机构和投资人的助力，但切斯基在公开博文中深情致谢的，依然是 Y Combinator 以及当初最早向他伸出援手的投资人（如 Paul Graham 等）。这种铭刻于心的认同感表明，成为第一个下注、第一个公开表达支持的人，具有不可替代的社交与声誉价值。当你公开为有潜力的人站台背书（Cosign），他们不仅会永远记住你的知遇之恩，更会形成强烈的示范效应，吸引更多顶尖人才主动向你聚拢。 - **数量级的人情记忆**：相比后期投入巨额资金的锦上添花，人们对最初在低谷或早期相信自己的人，记忆与感激程度往往高出数个数量级。 - **公开背书的价值放大**：“公开（publicly）”表达支持尤为关键，它不仅赋予被支持者宝贵的公信力资产，更确立了背书者发掘早期顶级人才的行业声誉。 - **正向网络效应**：成为优秀人才最早的公开支持者，会产生磁石般的吸引力，促使未来更多处于起步阶段的优秀建设者慕名而来。 ```mermaid flowchart TD accTitle: 早期相信者的声誉复利飞轮 accDescr: 展示成为最早支持者如何通过公开背书形成长期互惠与人才吸引的正向循环。 subgraph EarlyStage ["早期下注阶段"] A["发掘早期潜力人才 / 早期产品"] B["成为最初相信者并公开背书 (Public Cosign)"] A --> B end subgraph LongTermGain ["长期收益阶段"] C["人才取得巨大成功 / 公司走向上市"] D["获得高出数量级的铭记与回报<br/>(如优先投资机会、公开致谢)"] B --> C C --> D end subgraph Flywheel ["声誉飞轮效应"] E["建立卓越的'伯乐'声誉<br/>吸引更多顶级潜力人才主动寻求支持"] D --> E E --> A end ``` ### [49:32](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2972s) 结语与致谢 在探讨完“成为最初相信者”的深远意义与声誉复利效应之后，主持人 Erik Torenberg 与嘉宾们以这一充满启发性的核心观点为本期对谈画上了圆满的句号。大家互相致谢并总结，强调了无论在硅谷还是更广泛的职业生态中，尽早发现人才、真诚公开支持他人，不仅是对卓越建设者的赋能，更是塑造自身不可替代声誉资产的终极方式。 ## Notable quotes > [00:00](https://www.youtube.com/watch?v=-ywZlfznTa4&t=0s) 在这个行业、甚至在整个世界上，最重要的能力就是先于市场发现事物的能力。 > [00:00](https://www.youtube.com/watch?v=-ywZlfznTa4&t=0s) 做到足够优秀，以至于人们借由你来做生意 — Naval Ravikant > [00:53](https://www.youtube.com/watch?v=-ywZlfznTa4&t=53s) 去斯坦福大学念书的人，和被 Patrick Collison、Elon Musk 以及 Marc Andreessen 关注的人，哪一个更令人印象深刻？ — Erik Torenberg > [00:53](https://www.youtube.com/watch?v=-ywZlfznTa4&t=53s) 硅谷的迷人之处在于“你永远不知道下一个改变行业的大人物会是谁”，他们可能只有 18 岁，或者来自毫无名气的高校。 > [08:37](https://www.youtube.com/watch?v=-ywZlfznTa4&t=517s) 如果你想进入这个领域，就去这些公司 — Packy McCormick > [24:32](https://www.youtube.com/watch?v=-ywZlfznTa4&t=1472s) 不怕练过一万种踢法的人，只怕把一种踢法练了一万次的人。 — Bruce Lee > [34:25](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2065s) 在 AI 使得冷启动外联和申请成本归零的时代，最极致的稀缺品变成了“人类真金白银的背书、人类的由衷信服与深度了解”。 > [40:51](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2451s) 一切都会成功（It's all gonna work） — Eric Vishria > [48:09](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2889s) 既然这个人是我的‘Alpha’（独家超额优势），我知道他极具潜力，我为什么要把他公之于众？私藏起来岂不是对我最有利？ — David Booth > [48:09](https://www.youtube.com/watch?v=-ywZlfznTa4&t=2889s) 我今天能拿到这笔融资，正是因为两年前 Olivia 为我做了 Co-sign ## Mentioned a16z, Erik Torenberg, Josh Elman, Olivia Moore, David Booth, Cosign, Naval Ravikant, David Perell, X, Twitter, Patrick Collison, Elon Musk, Marc Andreessen, Zach Frankel, Russell Kaplan, Tesla, Cognition, Justine Moore, Stanford University, Packy McCormick, Crunchbase, PitchBook, SV Angel, Nikita Bier, Rise Awards, LinkedIn, Microsoft, Reed Hoffman, LinkedIn Jobs, Astra, MySpace, Charles Hudson, GitHub, AngelList, Wikipedia, Eric Vishria, Benchmark, Patrick O'Shaughnessy, School, tbh, Gas, Paul Graham, Sam Altman, Steve Jobs, Y Combinator, OpenAI, Ben Horowitz, Ali Ghodsi, Databricks, Ryan Hoover, Product Hunt, Dwarkesh Patel, Mark Zuckerberg, Michael Truell, Cursor, Leopold Aschenbrenner, Brian Chesky, Airbnb ## Video description a16z’s Erik Torenberg sits down with Josh Elman, Olivia Moore, and David Booth to introduce Cosign, a new product built around professional reputation and the people, companies, and products you’re willing to put your name behind. They unpack a simple idea at the heart of Silicon Valley: some of the most valuable professional signals aren’t credentials, but who believes in you. From the mentor who shaped your career to the colleague you’d work with anywhere or the young builder you think everyone should be watching, Cosign is an attempt to make those signals more visible and durable. They also discuss why human endorsements may become more valuable as AI makes outreach and information abundant, what existing professional networks get right and miss, and how making reputation more legible could help talented people get discovered earlier, find collaborators, and carry the work they’ve done behind the scenes into whatever they do next. Timestamps: 00:00 - Intro 00:53 - What Cosign Is & Why It Exists 07:24 - The Three Signals: Mentors, Investors & Cosigns 10:27 - Building the Anti-LinkedIn: Talent Discovery Redesigned 16:36 - Why Reputation Should Be Granular, Not Generic 22:52 - The Feed: Questions, Announcements & Signal Density 28:41 - The Meme Dynamic: Public Cosigns as Social Capital 36:50 - Existing Platforms: Why LinkedIn & X Fall Short 44:03 - Silicon Valley as Meritocracy & the Power of Being First 49:33 - Closing Thoughts Resources: Follow Josh Elman on X: https://x.com/joshelman Follow Olivia Moore on X: https://x.com/omooretweets Follow David Booth on X: https://x.com/david__booth Follow Cosign: https://x.com/cosign_build Read more about Cosign: https://www.a16z.news/how-silicon-valley-knows-its-people Stay Updated: If you enjoyed this episode, be sure to like, subscribe, and share with your friends! Find a16z on X: https://twitter.com/a16z Find a16z on LinkedIn: https://www.linkedin.com/company/a16z Listen to the a16z Show on Spotify: https://open.spotify.com/show/5bC65RDvs3oxnLyqqvkUYX Listen to the a16z Show on Apple Podcasts: https://podcasts.apple.com/us/podcast/a16z-podcast/id842818711 Follow our host: https://x.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details please see http://a16z.com/disclosures. [Source](https://www.youtube.com/watch?v=-ywZlfznTa4)

[0:00] 在这个行业、甚至在整个世界上，最重要的能力就是先于市场发现事物的能力。 [0:00] Naval Ravikant: 做到足够优秀，以至于人们借由你来做生意 [0:53] Erik Torenberg: 去斯坦福大学念书的人，和被 Patrick Collison、Elon Musk 以及 Marc Andreessen 关注的人，哪一个更令人印象深刻？ [0:53] 硅谷的迷人之处在于“你永远不知道下一个改变行业的大人物会是谁”，他们可能只有 18 岁，或者来自毫无名气的高校。 [8:37] Packy McCormick: 如果你想进入这个领域，就去这些公司 [24:32] Bruce Lee: 不怕练过一万种踢法的人，只怕把一种踢法练了一万次的人。 [34:25] 在 AI 使得冷启动外联和申请成本归零的时代，最极致的稀缺品变成了“人类真金白银的背书、人类的由衷信服与深度了解”。 [40:51] Eric Vishria: 一切都会成功（It's all gonna work） [48:09] David Booth: 既然这个人是我的‘Alpha’（独家超额优势），我知道他极具潜力，我为什么要把他公之于众？私藏起来岂不是对我最有利？ [48:09] 我今天能拿到这笔融资，正是因为两年前 Olivia 为我做了 Co-sign

[0:00] Naval Ravikant: be so good that people do business through you [11:05] David Booth: be so good that people would do business through you [12:52] Erik Torenberg: I would fund them if they start a company. [12:52] Erik Torenberg: I would hire them if they consider leaving their current role. [12:52] Erik Torenberg: I would like to meet them. [18:36] Erik Torenberg: It's not like... Mark and Andre wrote a blog post about me joining the firm—like, that's pretty dope. I want that on my... I want just the first thing when you think about me is to like think about Marc Andreessen's professional accolades. [24:32] Bruce Lee: Fear not the guy who has a thousand moves, but the guy who does one move a thousand times. [32:28] Olivia Moore: I don't think I know them. Maybe I met them at a party. [40:56] Eric Vishria: it's all gonna work. [40:56] That's a huge win. [44:16] David Booth: It was the ability to see which companies and which geographies, which people at those companies... it felt like it could be the LinkedIn for this sort of little niche of people who were ambitious in technology. [44:16] Erik Torenberg: the Nikita Bier of founder communities.

[0:00] The most important skill in this [0:01] industry or perhaps in the world is your [0:03] ability to identify things before the [0:05] market. [0:06] >> There are so many people in the world [0:07] who we want to meet who also want to [0:08] meet us and it's just an information [0:10] problem. What's more impressive to you? [0:11] Someone who went to Stanford or someone [0:13] who is followed by Patrick Collison, [0:15] Elon Musk, Mark Andre. [0:17] >> A lot of people in Silicon Valley have a [0:19] pretty high bar for who they're willing [0:20] to publicly endorse. A lot of what we do [0:23] it's finding awesome people and [0:25] connecting them to our companies. [0:26] otherwise kind of getting them in the [0:28] A16Z network [0:29] >> and it allows us to really extend all [0:32] that network and all that value and all [0:33] that trust [music] and really that [0:35] reputation that transfers and builds [0:37] over time. [0:37] >> There's this novel quote I love be so [0:39] good that people do business through [0:40] you. In fact, if I co-sign two people, [0:42] they might be able to meet and otherwise [0:44] build something on a way that I enabled [0:46] without even having to know about it. [0:47] >> But [music] it's always hard to match [0:48] the needs and the skills at exactly the [0:50] right time. What we've enabled is this. [0:55] We're super excited to announce cosign a [0:58] product that is around professional [1:00] reputation for the startup community and [1:03] really just trying to be the most [1:04] comprehensive directory of of companies [1:06] and people. And I want to start with a [1:08] story um which is my friend David Prell [1:12] uh told me he said my Twitter account is [1:15] more valuable to me than my college [1:16] education. Uh I would you know even [1:18] though I paid you know quart million [1:20] whatever it is you know I would have my [1:21] Twitter has given me more value to my [1:23] career and also personally. Now we of [1:26] course just launched our own school. So [1:27] we you know are big fans of uh of higher [1:30] education but I I I thought that story [1:33] was profound because of course you know [1:34] Twitter has been free for so long you [1:35] know um and um this idea what he got [1:38] behind it is of course college is a [1:40] bundle of you know education network and [1:44] credential and I think [1:45] >> and social life. [1:46] >> Yes. [laughter] Yes. Of course. Uh yeah [1:49] some of that is hard to recreate on on [1:51] Twitter but yeah exactly. Um, and so the [1:54] education, you know, you learn from [1:55] having conversations, the the network, [1:57] of course, you you meet people. Um, you [1:59] discover people. Um, people kind of is [2:01] proof of work of like if they're [2:02] interesting, if they're funny, etc. If [2:04] they build something cool, you can [2:05] discover them. But I think the [2:06] credential thing is really interesting [2:07] because here's a thought experiment. [2:09] What's more impressive to you? Someone [2:11] who went to Stanford or someone who is [2:13] followed by Patrick Collison, Elon Musk, [2:16] you know, Mark Andre, etc. People would [2:19] say, I think most people would say the [2:20] latter, right? And so this kind of like [2:22] peer-to-peer credential. And so what [2:25] that is, but you know implicitly, [2:28] although people also do this explicitly, [2:30] is a cosign. It's a it's an endorsement [2:32] that this person is worth tracking. It's [2:34] a public signal that this person is [2:36] worth following. And we think there are [2:39] lots of other ways to express that that [2:42] signal that that um and create this kind [2:44] of like reputational um you know benefit [2:47] that these people get which is really an [2:48] exhaust of this person is not even [2:50] trying to give them reputational [2:51] benefit. They're just trying to read [2:52] what they write and the exhaust is this [2:55] endorsement. And this endorsement if [2:57] you're followed by these people now your [2:59] doors are open. People want to re also [3:01] read what you say. They want to work [3:02] with you. They want to invest in you. [3:04] they they want to, you know, um have you [3:06] part of their team and also there's [3:08] benefit so there's benefit to the [3:09] receiver but there's also benefit to the [3:11] giver too because if you're the first [3:13] person to discover someone well that's [3:15] kind of how Silicon Valley operates like [3:16] you're more you know you're going to be [3:18] able to hire them quicker you'll be able [3:20] to fund them quicker and the cool thing [3:21] about Silicon Valley is you don't know [3:22] who the next big person is going to be [3:24] always and sometimes they're 18 or you [3:26] know or come from you'll come from an [3:27] abrand school and so we're kind of all [3:30] in this you know a little bit of a race [3:32] to discover talent um before anybody [3:34] else and get the more that we get like [3:37] Zack Franle's done an amazing job right [3:39] of bu so many people have done amazing [3:40] of building a reputation for discovering [3:42] people and so talent the benefit they [3:44] get is talent thinks oh I can get king [3:46] made by them like if Josh Alman or [3:48] Olivia invest in my company that's a [3:50] signal to the market and angel investing [3:52] or or just investing uh from venture [3:54] firm that's one way to express an a [3:57] co-sign or a signal but a lot of people [4:00] aren't founders and a lot of people [4:01] aren't investors and so we think there [4:03] lots of other ways to express this this [4:05] signal. [4:06] >> Really curious as you've been building [4:08] cosign like what are the kinds of [4:10] endorsements you hope to get from the [4:11] platform. [4:12] >> Yeah. So we want on every profile to [4:15] have at least the the three one is who [4:18] shaped your career. Um because we we [4:20] think this sort of like mentor uh econom [4:23] this what Silicon Valley runs on and it [4:25] also just it helps both the you know if [4:28] I was shaped if my you know if Josh Elma [4:30] is my mentor that means that he is [4:32] putting which was true actually [4:33] [laughter] [4:34] you know just happens to be true you [4:36] know said I was a young person Josh [4:37] spent a bunch of time with me and that's [4:40] a signal to other people that I'm you [4:41] know I'm worth other people spending [4:43] time with too because if you respect [4:44] Josh you respect me by association and [4:46] it's also again the benefit to you to [4:48] discover talent ly um to be known people [4:50] want to work with you if you help them. [4:52] So one has shaped my career. Two is [4:54] would work with anywhere or in the [4:56] trenches with um because you know it's [4:58] hard to tell from existing platforms [5:00] like you as you were saying the other [5:01] night who are you really in the trenches [5:03] with like who have you really built uh [5:05] you know bonds with who would you you [5:06] know go to war with who would you like [5:08] no matter what you're doing you'd want [5:09] to work with them that's a much more [5:11] granular signal than do I follow them [5:13] which could be any number of things. Um [5:15] and then third is uh person to watch [5:18] like who do you think is going to um [5:21] really be emergent like ascend it. Um [5:23] and we rank order these too. So you know [5:27] platforms just started you know most of [5:28] it's fair game but if if you just you [5:30] know in the same way like you know on to [5:33] build a great investor profile it takes [5:34] like uh you know many years but then [5:36] you've got credit for doing so. We we [5:38] think you do that around people too. So [5:39] you can say hey I think like I think [5:41] Olivia is going to be one of the greats. [5:43] I'm going to like, you know, I think [5:44] she's a person to watch. It's actually [5:45] funny. I did this experiment seven years [5:47] ago where I emailed a few hundred [5:48] people. You were definitely one of them. [5:49] And I asked them for their person person [5:51] to watch. I was just checking today. One [5:52] of them was Russell Kaplan who's now the [5:54] president at Cognition. And he was like [5:56] a 22-year-old just engineer at Tesla. [5:58] And so there's actually signal in that [6:00] too. Um, so those are some of the ones. [6:03] We're also excited for people to do it [6:04] for companies too. like what are the [6:06] companies that you think right you know [6:07] at seed but you know are you know going [6:09] to be great because of course if you're [6:11] able to you can express your conviction [6:12] with uh your you know money but not [6:14] everyone's an investor this is almost [6:16] like fantasy investing or social capital [6:18] investing [6:18] >> yeah I love this idea I feel like I have [6:21] long felt this but as I have invested [6:24] now for almost a decade I believe it [6:26] more and more every week every day every [6:28] month is like the most valuable data in [6:31] all of Silicon Valley is like who has [6:33] conviction and [6:35] uh and who can give you kind of a read a [6:37] real read on a person for so many [6:39] decisions like who do you invest in who [6:41] do you hire what company do you join uh [6:44] and there's such information asymmetry [6:47] right now like our job investing every [6:49] day we are pinging people saying hey you [6:53] know do you know anyone who worked with [6:54] this person and we do it every day and [6:56] we still I would say like have to work [6:58] really hard to find that signal and so [7:00] how can someone who is maybe evaluating [7:02] between five startup job offers [7:06] um get that level of you just can't [7:08] especially if you're like an entry-level [7:09] engineer and yet that could be the [7:11] difference between joining like a [7:12] complete rocket ship and like a company [7:13] that isn't doing as well. Um and I know [7:17] the data has to be handled carefully [7:18] around who says what about who, but I I [7:20] think it it will be and there's like a [7:23] massive massive opportunity to add so [7:25] much value to people's lives and [7:26] careers. [7:28] >> No, talk a bit about um because you [7:30] think a lot about talent discovery. um [7:32] talk about the newsletter you did [7:33] because I think as an interest [7:34] incredible [7:35] >> stuff. Yeah. So um my twin sister [7:37] Justine also works here at Andre on our [7:39] infrastructure team. Um we have both [7:42] been in venture for basically a decade [7:44] now and at our very first venture role [7:46] we started a newsletter that we ran [7:48] every Sunday for I think like six and a [7:51] half or seven years. [7:53] >> Subscriber. [7:54] >> Thank you so much. Um it was basically [7:57] just aggregating startup job postings [7:59] and the the thesis was like we both went [8:02] to Stanford, we ran a startup incubator [8:04] at Stanford and still before we were [8:07] working at a venture firm, we had no way [8:09] to differentiate between all these [8:10] startups and founders of like who is [8:13] legitimate, who's not, who's the most [8:15] exciting. And so we were like, okay, we [8:16] have some inside information now. Let's [8:18] like curate a list of like the companies [8:20] that that we think are worth joining for [8:22] young people. Um, and it really it built [8:26] up a real community because I think [8:29] people appreciated it because you can't [8:30] get that information outside of Silicon [8:32] Valley. Like we had a a network of like [8:35] scouts on campuses that Reddit and some [8:37] of those people are now founders that [8:38] Andre has invested in. They're all [8:40] running like fantastic companies and in [8:42] fantastic places. So, um, that was like [8:46] the less scalable version of this kind [8:49] of network. And I'm I'm excited for more [8:51] people to have the ability to kind of [8:53] put their name behind ideas. [8:55] >> One thing I'd have is that as it gets [8:57] easier and easier to start companies, as [8:59] it gets noisier and noisier out there, [9:01] um it's actually getting harder for [9:02] people to do what you're describing. [9:04] Exactly. So this kind of signal is more [9:05] important than ever. Um if you think [9:07] about uh the average student graduating [9:09] out of sample today, they're looking at [9:10] the startup market and they have to rely [9:12] on the signals that are given by firms [9:15] who are investing but even before that [9:16] that are given by people like you and [9:18] like hundreds of others we hope to [9:19] empower here. Um an example I love to to [9:22] point to on the site when people go and [9:24] you've got people like Py McCormack who [9:25] I think has has always had a lot of very [9:26] good signal around the deep tech and the [9:28] the vertical integrators in particular [9:29] and Paky's got 37 companies that he's [9:31] chosen. He says these are the companies [9:33] if you want to get into one into the [9:34] space. Um, and there's, you know, many, [9:36] many more behind them, self included. [9:38] Uh, so there was like you're in a way [9:40] he's co-signing some of his reputation [9:42] against that company. He's saying like, [9:43] I trust this company. I'm putting my [9:44] money where my mouth is. I'm writing [9:46] about them. And I think that if you [9:48] wanted to start your career off there, [9:50] um, so it becomes a sort of living [9:51] directory of companies and of people [9:53] that are worth paying attention to [9:54] >> and free, you know, crunching base, [9:56] pitchbook, you know, cost [9:57] [clears throat] money, etc. They're [9:58] great products, they have to be [9:59] businesses, but we have the luxury of [10:00] being able to just really offer this for [10:02] the startup community. And so we we want [10:04] it to be the most valuable place where [10:05] people can see you know who invested in [10:07] what you know what is uh you know [10:09] Olivia's uh you know uh set of deals [10:12] that she's done um you know what is the [10:14] sort of most reliable you know company [10:16] information on like who are the [10:17] co-founders who are employees what are [10:18] the products they shipped and one thing [10:20] we've decided is that you know no [10:22] product can be all things to all people [10:24] and there are trade-offs and we're this [10:26] is just going to be positive so this is [10:28] going to be a place where people brag [10:30] it's going to be accurate but um you [10:32] know [10:33] How often do you go to uh you know [10:35] someone's uh LinkedIn and or text [10:37] someone and be like hey you're connected [10:38] to someone on LinkedIn like can you tell [10:40] me about it? You're like ah I don't [10:41] really know them. You like we're going [10:42] to have a much more granular information [10:44] to make sure that it's um there's real [10:46] signal there. Um but uh that is you know [10:49] one product decision that we've we've [10:50] made. David, I'm I'm this is something [10:53] that we experimented with seven years [10:54] ago and decided not to pursue because we [10:56] got busy with other stuff, but I'm [10:58] curious what has always struck you about [11:00] this idea in terms of the the need for [11:02] it or you know what we're really trying [11:04] to do here? [11:05] >> One thing I've always believed is um the [11:08] most important skill in this town or in [11:10] this industry or perhaps in the world is [11:12] your ability to identify things before [11:15] the market. And those things if you're [11:17] an investor, it's companies. If you're a [11:18] hiring manager, if you're a founder, [11:20] it's people. Um, sometimes it's [11:22] products. If you are, you know, in [11:23] charge of procurement for, you know, a [11:25] growing company, maybe you need to [11:26] identify which product is better for the [11:28] market. Um, but there's no way of [11:29] recording that. There's no there's no [11:30] hipster leaderboard that says this [11:32] person found this coffee shop that, um, [11:35] for angel investors, of course, um, they [11:37] want to be non-conensus and right, and [11:39] they're rewarded for doing so because I [11:41] can invest in your preed round. If I'm [11:42] right, then it's lucrative for me. Um, [11:45] one of the things that's always drawn me [11:46] back was the idea that I can't invest in [11:48] my my personal social capital in you. I [11:51] am very willing to make an introduction [11:52] for you. Like if I really really believe [11:54] in someone and no one else has heard of [11:55] them, I can make an introduction to [11:57] them, you know, Josh Ilman, for example, [11:59] um, and he will hopefully take that [12:00] introduction because he's see sees me [12:03] putting my reputation against that [12:04] introduction. He has no idea who this [12:06] person is, but he knows that David [12:07] thinks highly enough of this person to [12:09] take a chance on him. To me, the the [12:11] thing that stood out the first time [12:12] around and the thing that we're trying [12:13] to really lean on coming back is um like [12:17] when I co-sign something, I take it [12:18] seriously. I want that person to be able [12:20] to take the cosign I've given given them [12:21] and they might be able to go straight to [12:22] Josh and they say, "Hey, David cosign me [12:25] and Josh might see that he might reply [12:27] to that cold DM." Um and in fact, that [12:30] uh sort of lubricates the the wheels of [12:31] social capital that should you could [12:34] empower a lot more people to achieve a [12:36] lot more. There's this novel quote I [12:38] love from a long time ago. be so good [12:39] that people would do business through [12:40] you. In fact, if I cosign two people, [12:42] they might be able to, you know, meet [12:44] and and otherwise build something or [12:46] otherwise hire each other or invest in [12:47] each other in a way that I I enabled [12:49] without even having to to know about it, [12:51] which is very exciting. [12:52] >> Yeah. I um it's really interesting [12:55] related to that, I want to talk about [12:56] more features of the product. I was [12:58] talking about sort of the more granular [12:59] ways of expressing reputation. One of [13:01] the best events I ever went to uh was [13:03] this SV Angel conference in like 2018. [13:06] They they I think they do once a year or [13:07] something like that. And be every event [13:10] should do this. But before I I went to [13:11] the event, they had this list of 150 [13:13] people that they sent out and it was you [13:14] know all of it was you know everybody [13:16] was a you know baller and um they said [13:19] swipe who you want to meet uh who you [13:21] want to intro to and I swiped on almost [13:24] everybody. um you know it was up and [13:26] coming you remember Josh um and eager to [13:28] eager to meet great people and basically [13:30] if they swiped right on me then we would [13:32] get automatic intro and it would be like [13:34] you guys have full context you know you [13:35] should chat and um I then ended up [13:38] meeting with most of those people and [13:39] they became like really like so that [13:41] event of course I couldn't have met you [13:43] know 40 people at an at an event and had [13:45] deep conversations with them but that [13:47] event provided the context for us to you [13:50] know become friends and collaborators [13:52] and I say that to say that there's [13:54] there's so many people in the world who [13:55] we want to meet who also want to meet us [13:57] >> and it's just an information problem and [13:59] if you can reduce the friction to make [14:02] that introduction because right now it's [14:04] like I have to DM them but I have to [14:05] think about something to say are they [14:06] interested I don't know but this is just [14:07] sort of passive like yeah here are [14:09] people I would meet now what we've [14:10] enabled is this privately signal it's [14:13] actually a bit more uh granular so it's [14:15] not just I want to meet them though [14:16] there is that but it's also I would fund [14:19] them if they start a company or I would [14:21] hire them if they consider leaving I I [14:23] can't there's there's a lot of people [14:24] who I would hire, but I can't like reach [14:26] out to all of them and also might be [14:27] inappropriate and what like I'm too busy [14:29] and I would only want to talk to them if [14:30] they want to join me or whatever. And so [14:33] um having more granular granularity [14:35] around professional intent um passively [14:39] there's no low no risk of like cuz [14:41] there's some cost in sending a DM. I [14:42] don't know you might feel rejected or [14:44] it's time or whatever it is. And if they [14:46] also express interest then it's an [14:48] introduction with high intent I think [14:50] can unlock just a lot of uh you know [14:52] good stuff. Yeah, that conference [14:54] example is so interesting because I feel [14:55] like the other takeaway for me is like [14:58] we can all be people matchers now with a [15:00] product like full sign. Like I think [15:02] about in in all of our jobs, we meet so [15:04] many people every day. I often meet [15:06] someone I'm like they are awesome. They [15:08] have this really specific skill set. [15:10] Then I meet someone else three months [15:12] later who's looking for that skill set [15:14] and I have to figure out like remember [15:16] this person, check if they still want [15:18] the intro, make the intro and so like [15:20] that store of information and somewhere [15:22] that's more persistent is and can kind [15:25] of work across time is really really [15:26] interesting. [15:27] >> I mean I agree with that completely. You [15:28] know so much of what we like to do both [15:31] as investors and in people around [15:33] Silicon Valley is make these connections [15:35] but it's always hard to match the needs [15:37] and the skills at exactly the right [15:39] time. And remember that and being able [15:40] to codify this. Here's the people I [15:42] think are truly great designers or [15:44] here's the people I think are truly [15:46] great PR people who can help you through [15:48] a crisis or here's the people that have [15:50] really helped me through, you know, a [15:53] career decision that I was making and [15:54] are the mentors that I go to. Being able [15:56] to actually endorse that now has it [15:59] available so that when you're looking [16:00] for that, you don't just have to go, [16:02] Josh, [16:03] >> who can help me think through this [16:04] decision. And I have to go through and [16:06] see who have I recently talked to, who's [16:08] fresh, top of mind that I can think of. [16:10] But I can really go back and be able to [16:11] say, "Hey Josh, I'm looking through [16:13] those cosigns and you said that this [16:15] person really helped you in this one [16:16] situation. Like, do you think they'd be [16:18] open to helping me?" And it allows us to [16:20] really extend all that network and all [16:22] that value and all that trust. And yet [16:25] still is a way that you're trusting me [16:27] to try to make that introduction. Like [16:29] it's not that any of us want to just [16:30] say, "Hey, you know, everything's open [16:32] to everybody." it's really still part of [16:34] the trusted community and really that [16:37] that reputation that transfers and [16:38] builds over time. [16:39] >> So building on something Eric said the [16:41] private signal um is one core piece of [16:44] it and I think what you're talking about [16:45] here is in a way the public signal. So [16:46] we've built the two of them the private [16:48] signal um I mean honestly the the [16:50] product is so fun you can go rabbit [16:51] holeing around. It's a bit like [16:52] Wikipedia clicking from link to link to [16:54] person to profile to company to investor [16:56] and back again. And as you go, you're [16:58] collecting um maybe it's it's Pinterest [17:01] that I'm going to put this person on my [17:02] personal private list of people that I I [17:04] want to, you know, work with in the [17:05] future. And and I'm going to put this [17:06] person on my public list of designers [17:08] that I I rate. And every time a founder [17:10] comes to me and they say, "David, who I [17:12] need I need a new designer. I need a a [17:14] growth guy. I need an engineer." Um I [17:16] can say, "Well, do I consult my private [17:18] signals or do I send them my public [17:20] list?" And hopefully everyone has this [17:22] sort of habit of if I really believe in [17:24] someone, I'm putting them on my private [17:25] list. It's actually utility both ways. [17:26] It's telling them that I believe in [17:28] them. It's telling their network that I [17:30] believe in them. It's also something I [17:31] can use. I can turn around. I can give [17:33] to a founder or someone I respect. I can [17:35] kind of shortcut that work that you're [17:36] saying you you you do you have to do it [17:38] every time someone asks a question. [17:40] >> What's so interesting about that to me [17:41] as well is like there's this often [17:43] timing discrepancy that happens in [17:45] networks where I'm like, "Oh my gosh, [17:47] you know, I know this person has been [17:49] trying to get their startup off the [17:50] ground. this other startup would like [17:52] love to, you know, merge with them or [17:54] welcome them to the team, something like [17:55] if the timing is right. But if I go to [17:57] that founder and I'm like, hey, you [17:59] know, are you thinking about exiting [18:00] your company? That's like a big risk. It [18:03] could be like offensive or insulting or [18:05] maybe the company is working and they [18:06] want to keep going on it. And so the [18:08] idea of just like having this two-sided [18:12] network where there's information on [18:13] both sides and they can match when the [18:15] time is right when they indicate I think [18:17] is like really really really great on [18:19] both hires, acquisitions, like job [18:22] changes, tons of things. [18:24] >> Totally. [18:24] >> Yeah, that's so fascinating. The company [18:26] to company I didn't even think about [18:27] like I would buy stuff. [laughter] [18:30] >> Yeah. [18:31] >> We'll get the product features tomorrow. [18:32] >> Exactly. [18:33] >> We just extended the road [18:34] >> path. Yeah, that happens. the there are [18:37] a couple things I want to add to that. [18:38] So, you know, it's interesting. You [18:40] know, we thought a lot about, you know, [18:41] should there be a feed, you know, if so, [18:42] what is on the feed? Is it free form, [18:44] etc. And Josh helped us really nail down [18:46] sort of the focus of the feed to be [18:48] around questions and announcements. And, [18:51] you know, announcement is so interesting [18:52] because um we think a way to [18:54] differentiate there is to basically make [18:56] reputation durable. A lot of reputation [18:58] is uh built on social media, but it's um [19:01] it's ephemeral. So when Josh joined the [19:03] firm, you know, he wanted me to focus on [19:05] other people, but you're such a good [19:06] example. Uh, a lot of people came out of [19:08] the woodwork, um, who've been working [19:10] with him for a long time, um, but who [19:12] maybe haven't, you know, publicly said [19:13] anything recently because they didn't [19:14] need to and said, "Wow, you know, I [19:16] really endorse Josh. You should work [19:17] with Josh. Josh helped me in my career." [19:18] You know, I said that, Nikita said that, [19:20] a bunch of other people said that. And [19:22] then, you know, the next day just [19:24] happened and and some people saw that, [19:26] some people didn't see it. But that type [19:28] of stuff is so valuable and should exist [19:29] on someone's profile like the first time [19:31] you're looking them up. And so for us, [19:33] any uh announcement is going to be tied [19:35] to their sort of, you know, durable [19:37] reputation. Um, and so when companies [19:40] fund raise, when people switch jobs, [19:41] these are opportunities for for co-signs [19:43] for for endorsements, etc. And then on [19:45] the question side, it's really this, you [19:47] know, this what we were talking about [19:49] earlier is like who's the best comms [19:50] person for this or who's the best angel [19:52] investor in fintech or or you know, who [19:54] are people I should work with for this [19:55] or this kind of like um you know, very [19:58] specific uh you know question of of some [20:00] sort of person or company or product [20:02] you're looking to either surface to you [20:04] or or learn more about and um in that [20:07] way will also build durable reputation [20:09] because you get added to a list through [20:10] that. No, I I think this idea of [20:12] bringing those ephemeral moments and [20:14] those ephemeral conversations that [20:15] happen on Twitter and other social X and [20:18] other social media forums all the time [20:20] are like there's so much that just [20:22] happens and it's wonderful in the moment [20:24] and then it gets lost. And I think this [20:26] idea of of building a profile, not a [20:28] profile that is just what you say about [20:30] yourself, but really that lets other [20:33] people, you know, be positive and help, [20:35] you know, shine on your strengths, I [20:37] think is really really powerful. and and [20:39] you know I I see a lot of these for you [20:41] and the great work you're doing here. I [20:42] see a lot of these for Olivia and you [20:44] know David like we get these comments [20:47] and and it's just so special and you [20:49] feel so lucky to have people you know [20:51] that you've helped in the past be able [20:52] to say those things about you now but [20:54] then again it just gets lost and and I [20:56] think this idea of tracking those [20:59] co-signs and really making it feel like [21:00] this fundamental network of the the web [21:03] that we are of Silicon Valley of the [21:05] tech industry I'll give you really [21:07] powerful. Do you know the meme where the [21:09] guy's like walking into a party at Stick [21:11] Figures and he's like they don't even [21:12] know that um like uh that's something [21:15] you get like a baller cosigner or [21:17] whatever and you're like it's not like [21:18] like um you know Mark and Dre wrote a [21:20] blog post about me joining the firm like [21:22] that's pretty dope. [laughter] [21:24] like I want that on my like I want just [21:26] the first thing when you think about me [21:27] is to like think about Mark and George [21:29] professional ladies and so yeah there [21:31] just all these moments that happen if we [21:33] can make them more legible uh you know [21:35] you can more accurately represent [21:37] someone's [21:38] >> honestly it wasn't until we actually [21:39] hooked up the internet fire hose to this [21:40] product that I realize how many of those [21:42] moments are I mean you you look at the [21:44] the main feed and every day there is [21:45] fundraising announcements there's talent [21:47] transitions uh thanks to our good [21:49] friends at X we're we're plugged into a [21:50] lot of those and we're we're taking like [21:52] whenever somebody quote tweets and [21:53] announce announcement or uh replies and [21:55] it's a it's a we you're filtering them. [21:57] It's a genuinely thoughtful positive [21:58] comment your point earlier about [21:59] positivity. um those are pulled through [22:01] and it's and it's yeah it's really [22:03] coming from a person and it's on the [22:05] profile the person comes from and it [22:06] goes to a person [22:07] >> to and and the reason why it's all [22:09] positive is because products have to win [22:11] on some dimension and this product is [22:13] going to win on people um using it to [22:16] show off like to show off in a good way [22:18] like this is what I'm proud of like this [22:20] represents the like the full [22:22] manifestation of what I'm capable of on [22:24] a person level and on a company level. [22:26] >> Yeah. [22:27] >> And and I really think it's about it's [22:28] about celebration. I mean like so much [22:30] of what we do there's so much hard work [22:32] that goes on into making every single [22:34] product that happens. It's a toil of [22:36] somebody's love, their idea, the coding [22:38] they wrote or maybe that they worked on [22:40] with their agent now. But like putting [22:42] that out in the world is it's it's a [22:44] very vulnerable moment and it's very [22:46] like do you like what I've tried to [22:48] make? Do you you know are you going to [22:50] use it? And I think, you know, we need [22:52] to celebrate that effort and what goes [22:54] into those moments and what goes into [22:56] all the hard work that we do as opposed [22:58] to, you know, there's there's plenty of [23:00] places for critique and we see all that [23:02] all the time, too. But I think that [23:04] doesn't need to be recorded for [23:05] posterity in the same way. [23:06] >> And one thing we're talking offline is [23:08] interesting is a lot of, you know, [23:09] there's a lot of um awards and a lot of [23:11] shine on founders. Yeah. [23:13] >> Uh and you know, justify so obviously. [23:15] Um and also there's a lot, you know, a [23:17] lot of other people build the companies [23:19] too. Um, and a lot more people are [23:20] critical to those companies and they're, [23:22] you know, unknown uh or or not [23:23] celebrated to the same degree and this [23:25] is an opportunity to to do that as well. [23:26] >> Totally. I think even just positive [23:29] signal is going to be incredibly [23:31] informative cuz I think a lot of people [23:33] in Silicon Valley have a pretty high bar [23:35] for who they're willing to publicly [23:37] endorse in in any way. Like when I [23:39] >> think about like the number of intro [23:41] requests that come in and how many of [23:42] those do I actually bridge, like I I [23:44] think it's [23:45] >> cuz if you Yeah. Have you endorsed [23:46] someone who's not great? That reflects [23:47] poorly onact [clears throat] the natural [23:49] like reputation sin mechanism. [23:51] >> And there's a whole thing about, you [23:52] know, do you share something that you [23:53] haven't even played with? You know, [23:54] there's a whole, you know, have you even [23:55] read the article before you retweet it. [23:57] And I think I think the key is like [23:59] you'll get to know pretty quickly who's [24:01] just spouting things and is, you know, [24:03] celebrating everything that could be [24:04] celebrated versus, you know what, I've [24:06] actually played with this. I've spent [24:07] time with it. I've heard the backstory. [24:09] I've thought about the backstory. And I [24:11] just want to double down on one comment [24:12] you said about this going beyond the [24:14] founders. Like I I think you know the [24:16] founders who take that risk to start a [24:18] company from nothing. That's such an [24:19] incredible thing. But it takes a village [24:21] to build every one of these companies. [24:23] It takes so many great designers and [24:25] engineers and marketers and product [24:28] managers and we um you know finance and [24:32] like there's so many parts of every [24:34] single company that have to get built [24:36] and celebrating that work and when that [24:39] happens from a specific team or a [24:40] specific individual and be able to have [24:42] that be part of this kind of endorsement [24:45] and co-signing I think is going to be [24:46] incredibly powerful beyond just you know [24:49] the people that we talk about most of [24:50] the time. want tell a story or perhaps [24:53] nudge Eric to tell a story. One of the [24:55] first times I actually heard of Eric uh [24:56] and then went on to meet Eric was the [24:58] Rise Awards and the Rise Awards 2015 16 [25:00] thereabouts. So over a decade ago now [25:02] and it was around this thesis that um [25:04] you know farmers get a lot of publicity [25:06] a lot of credit but it's actually the [25:07] people who are the operators behind the [25:08] scenes and so the principle do you want [25:10] to tell the story what were the rise? [25:12] >> Yeah. So the the RISE awards were this [25:14] uh series of awards meant to be for [25:16] everybody else uh for the engineers, the [25:18] designers, the product. I believe Josh [25:19] was a judge uh you know many others and [25:21] we had thousands of of submissions and [25:23] um you know discovered and selected some [25:25] great people because of it and we're [25:27] going to bring that back to some [25:28] capacity. You know there's this what's [25:29] the phrase fear not the guy who has a [25:31] thousand moves but the guy who does one [25:33] move a thousand times. [laughter] [25:34] >> There's the samurai that's the kick a [25:37] thousand times. So, yeah, Rise Awards, [25:39] cosign, a lot of these ideas have been [25:41] in the ether for us for [25:42] >> in fact, the thing that stands out is a [25:44] lot of the people who were the Rise [25:45] Award recipients got to know each other [25:47] and built lifelong relationships through [25:49] the retreats they attended through the [25:51] groups that you added them to and [25:52] things. So, there's something around [25:53] that the connection too. It's it's [25:55] really interesting cuz um I feel like we [25:57] often meet people who are spitting out [25:58] of great companies that start their own [26:00] companies and often if you get really [26:03] deep like if you call everyone they've [26:05] ever worked with it's like wow this [26:07] person who was like I don't know [26:09] engineer number four actually was a huge [26:12] determinant in this company being [26:13] successful or not but that to your point [26:16] earlier about legibility that's not [26:18] something that they can like publicize. [26:20] They're not going to be like the face of [26:22] every article about the company. And and [26:24] yet there's like so many of these people [26:25] who [26:26] >> I think I would love to see them get to [26:28] leave that company if they go start a [26:30] new company, if they go to another one [26:31] and carry that accomplishment with them. [26:34] >> Totally. [26:34] >> And they can't really do that right now. [26:36] >> We're making the reputation market more [26:38] like efficient basically more liquid in [26:40] the sense of well both efficient and [26:42] liquid. So for example, there's some [26:43] people who have an amazing reputation [26:45] but there's not awareness. Not a lot of [26:47] people know that they have a great [26:48] reputation. Um, and then more liquid in [26:50] the sense of if it was known that they [26:52] reput great reputation, a lot of people [26:53] would want to hire them or fund them and [26:55] they should, you know, get those [26:56] opportunities as a as a result. And so [26:58] it's basically like trying to fix that [26:59] information asymmetry. Um, I want to [27:02] return to what you were saying earlier [27:03] because it's really interesting. It's [27:05] like, you know, a bunch of times a [27:06] candidate will, you know, or someone [27:08] looking for a new job will, you know, [27:09] sit you down and say, "Hey, what company [27:11] should I join?" And so, uh, and there's [27:13] a lot of people who wish they can do [27:14] that, but if you, you know, got your [27:15] list of of companies they should join, [27:16] it's kind of do business through you. [27:18] They can. It's like Olivia stamped these [27:20] companies. They can uh you know re reach [27:22] out and similarly um the the u and it'd [27:27] be cool also like for the ones you do [27:28] meet with um that you could just like [27:31] press a button and it would go to all [27:33] all your companies instead of having to [27:34] you know think through cuz there's [27:36] there's some that you might want to you [27:37] know share private just to keep to your [27:38] companies or or whatever it is. Um but [27:40] then also there are companies who want [27:42] to who are like hey I want a head of [27:44] comms. who in my network knows ahead of [27:46] like who should I even talk to and to [27:48] the extent that you know people are [27:49] co-signing heads of comms or or or this [27:51] kind of thing it can just like [27:53] facilitate much more you know matching. [27:55] >> Yeah totally. I feel like even for [27:57] investors like the most upto-date [28:00] profile of the companies they care about [28:02] is maybe if they tag them all in their X [28:04] bio and maybe if they put them on their [28:07] LinkedIn otherwise you have to go to the [28:08] firm's page and see if they connect them [28:10] to investments. And like I if if I could [28:13] and not ruin kind of the reach in the [28:15] audience, I would be tweeting about all [28:16] my companies all day every day [28:18] [laughter] [28:19] like promoting them. Go work for them. [28:20] They just did this amazing thing. But [28:22] you can't do that. [28:23] >> And I love the idea of a place where [28:25] like that kind of lives uh to see at any [28:28] time. [28:29] >> Yeah. Be shameless. [28:30] >> Yes, [laughter] [28:30] >> it is interesting. I mean, so um let's [28:33] talk about existing platforms for a [28:34] second. you X obviously, you know, such [28:36] a great platform and the exhaust of it [28:39] is so valuable to so many people, but [28:40] there's also times where like I'm like, [28:42] man, I wish I could just download my [28:43] follower. Like, I wish I could just like [28:45] where like, you know, a few thousand [28:47] people liked this thing about joining my [28:48] company. I wish I could just like see [28:49] this and why is it so difficult like or [28:52] it's not built for this use case of [28:54] like, you know, discovering people to [28:55] work for or just get more signal like, [28:56] okay, you follow me, but like, you know, [28:58] do you want to work for me? Do you want [28:59] to invest in me? I don't like what do [29:01] you want to what do we want to do here? [29:03] Um, and so I um, and then LinkedIn, you [29:06] know, it's such a valuable utility. [29:08] We're on it all the time and yet [29:10] everyone hates it. I'm I'm just [29:12] [laughter] I'm just We love [29:14] >> We love LinkedIn. We love LinkedIn. Josh [29:16] used to work there. LinkedIn, you know, [29:17] one of the most iconic products of of [29:18] all time. Mark famously defended it. I [29:20] remember when they when they got [29:21] acquired cuz like what only 26 billion? [29:24] It was like, you know, it's a it's a [29:25] great outcome. And it was iconic product [29:27] and and yet um, you know, people have [29:30] frustrations with it. So I'm I'm I'm [29:31] well maybe let's even even before the [29:33] feed I some people say the feed is cring [29:34] or whatever but even before the feed I [29:36] feel like people were like ah I wish [29:37] there was just more [29:38] >> more that we could do here. So why don't [29:40] you talk about that? [29:41] >> Yeah look I worked at LinkedIn back in [29:43] 2004 and five when we were just starting [29:45] to form the network and and Reed Hoffman [29:48] the founder used to talk about it very [29:49] much as like we have three stages of the [29:51] company like first is growth. We really [29:54] have to figure out how to make sure we [29:55] have the largest professional network [29:56] and give everybody value of just being [29:59] part of their network even if they have [30:00] nothing to do and actually for a lot of [30:02] users sign up make your profile don't do [30:05] anything else for a while unless [30:06] something comes to you was a perfectly [30:08] good usage then then we talked about [30:09] usage we really have to start to build [30:12] products on top of the network the the [30:14] real thing was the network is just the [30:16] substrate we need to go build a real job [30:18] service I helped launch the first [30:19] LinkedIn jobs we had to go build [30:20] services find the best PR professional [30:22] find those things um and and then we'll [30:24] build revenue. So first we so it was [30:26] kind of the the I called it gurr growth [30:28] usage revenue and and we really had to [30:30] figure out what those platforms and [30:32] products were going to be on top of the [30:33] network. Now the challenge is as you [30:35] built the network it was very single [30:37] signal are we connected and at the early [30:39] days the beauty of that was the idea [30:41] that we really know each other. I'll [30:43] vouch for Olivia and if David is asking [30:45] for an intro to um to me and Olivia says [30:49] David's really good um will you take [30:51] this intro? That's great. And it was [30:53] even better if you imagine this whole [30:54] table. David would ask Olivia for an [30:56] intro to say, "Hey, can you introduce um [31:00] me to Eric and then Olivia would say, [31:02] Josh, David's my really good friend. It [31:03] looks like you know Eric. Would you be [31:04] willing to pass him on?" And this idea [31:06] of this trusted graph, I'm so happy to [31:09] connect with you and I'll pass you on to [31:10] somebody else. Was the important [31:12] condition. But over time, it started to [31:15] fade from that that strength of [31:17] connection to like, "Yeah, we met once [31:18] at a party. We high-fived. We had a good [31:20] email exchange." And all of a sudden [31:21] then you you know everybody kind of got [31:23] into a collecting and the challenge with [31:25] collecting that network is it makes it [31:27] much harder to figure out where the real [31:28] signal is. Now on the other side it [31:31] still is the iconic representation of [31:33] who each of us is the professional [31:35] profile. But when we think about how to [31:36] use that professional profile it's only [31:38] really good if it adds value. Um, one of [31:41] the things I remember when we were first [31:42] building LinkedIn jobs was the goal was [31:45] to take a current job service where [31:46] someone just applies and gets a resume [31:48] and you see a job listing and go how do [31:50] we use the power of your professional [31:51] network to make it better? My very [31:52] favorite like Reed would push us a lot [31:54] and he'd say what is a social resume? [31:56] When a job application comes in and it [31:59] wasn't just here's who this person is, [32:01] it was here's who this person is and [32:02] here's who else in your company might [32:04] also know more about who the person is. [32:06] here's who else might have worked at [32:08] similar companies they' worked at in the [32:09] past or here's who knows people really [32:11] well who also worked at those companies. [32:13] So you wouldn't just get this flat [32:14] resume, you'd get a resume annotated by [32:17] who else can help you reach that person. [32:19] So that was like this big change to make [32:21] job applications socially enhanced and [32:24] then when you're looking at jobs you [32:25] might just see a flat job and like that [32:27] job looks really that company. We said, [32:28] "No, we're going to put the hiring [32:29] manager on. Their profile is going to be [32:31] on the job." All of a sudden, instead of [32:32] a faceless job that you're just applying [32:35] and throwing your resume over the wall. [32:36] Hey, you're going to get hired by this [32:38] person and then by the way, here's the [32:40] ways that you might have connections to [32:41] that person or you might have [32:43] connections to other people who work at [32:44] that company. So, it's again, how do you [32:46] use your network to do those things? And [32:48] so, those were these like beginning [32:50] epiphies of how to really rethink just [32:52] something as simple as a job application [32:54] and job site with this social network. [32:57] And I think again what was been missing [32:59] though is the set of things that happens [33:01] ephemerally through X like you were [33:03] talking about the set of things that [33:05] that we all have but we don't figure out [33:07] how to strengthen and put into LinkedIn [33:09] anymore. The true connection the reason [33:12] we were in the trenches together we [33:14] worked on this thing. You bailed me out [33:16] of a crisis. And when we can have that [33:18] that information in the network is going [33:20] to allow all these use cases to happen [33:23] so much better. And when you get to [33:24] focus just on kind of making sure the [33:26] network's right and getting all that [33:27] usage, I think cosign can really open up [33:29] so many opportunities. [33:30] >> Yeah, it feels like there's kind of like [33:32] a hierarchy almost of connections and [33:35] LinkedIn for good reason. Again, I feel [33:37] like we're all saying iconic product. [33:38] I'm an hourly active user [33:41] >> making a ton of revenue. They're clearly [33:42] doing fine, but LinkedIn treats [33:45] >> almost every kind of type of connection [33:47] the same, especially as products like [33:48] endorsements on LinkedIn haven't really [33:50] taken off. So, like for example, 90% of [33:53] the time when I say to someone, "Hey, I [33:54] saw you're connected to this person on [33:56] LinkedIn. Can you intro me?" I will get [33:58] the hm I don't think I know them. Maybe [34:01] I met them [laughter] at a party. I'm [34:02] not sure. So, there's like, "Do you [34:05] actually know the person? Did you work [34:06] at the same company? Did you work at the [34:08] same company at the same time? Were you [34:10] on the same team or adjacent teams that [34:12] work together? Did you manage them?" And [34:14] like all of those are treated as one [34:16] like connection or not connection now. [34:18] And I love the idea for cosign of being [34:20] able to like break that into the [34:22] granularity that it deserves cuz it's so [34:24] much more useful. [34:25] >> Seems like the greatest time in history [34:26] to be building products generally that [34:28] can accumulate context that was [34:30] otherwise unavailable. You say the [34:31] social CV. I love that idea. I think [34:34] Olivia in particular at the firm [34:35] thinking a lot about like the talent in [34:37] the social space and agents are an [34:39] interesting way to accumulate a lot of [34:41] context around a person. Uh I think a [34:43] lot of the things that we're building [34:44] here. Do you talk more about that? [34:45] There's there's a lot of [34:47] >> totally [34:47] >> what does the future hold? [34:48] >> So a lot of what we do it's both [34:50] actually our investing teams and our [34:51] operating teams is just finding awesome [34:54] people and investing in them connecting [34:56] them to our companies otherwise kind of [34:58] getting them in the A16Z network. And so [35:01] right now a lot of this is scraping [35:03] these existing you know profiles and [35:06] networks and then doing a lot of manual [35:08] work of going to see who actually is the [35:10] best strength of connection. How can we [35:11] make this happen? I think that that kind [35:14] of manual work that's truly external [35:17] signal based and it's um kind of binary [35:20] signal versus really granular signal [35:23] that's only going to get harder because [35:25] for example recently I posted on [35:27] LinkedIn I was like hey I'd love to help [35:28] my portfolio companies hire um if you [35:31] want to work at a startup fill out this [35:33] Google form [35:35] when it works it's almost harder because [35:38] I get 2,000 I get 2,000 results right [35:42] and and that's going to happen to every [35:43] person who posts a job or who posts any [35:45] kind of opportunity and and it's going [35:47] to be 2,000 now and it's going to be [35:49] 10,000 you know a month from now as [35:51] browser usage gets better and more [35:53] people use browser it's going to be [35:54] 100,000 and so it's like how do you even [35:57] me using Astra at like maximum strength [36:00] like how do you sift and sort and so in [36:02] an age I think of like endless [36:05] uh kind of outreach and interests with [36:08] AI like the scarce thing is like the [36:11] human endorsement or the human [36:13] conviction or the human knowledge. And [36:15] so that's where I get especially excited [36:18] because I think cosign is going to be [36:20] valuable hopefully so many people in the [36:22] community and also very valuable to the [36:24] work that we do and what we try to do [36:25] here. [36:25] >> And yeah, it's interesting. You know, [36:26] we've sort of gotten to a place where [36:28] there's only like broadcasting networks [36:30] to the whole world or like group chats [36:31] with like 50 people. There's not really [36:33] this like you know 10,000 or like a much [36:35] more vetted you know community and [36:37] that's where we are to start. um and [36:39] grow, you know, growing slowly and [36:40] basically where you put out a post and [36:42] you'll just be more confident in the [36:43] signal that that that comes in. Um going [36:46] back going back to LinkedIn for a second [36:48] on some some things that I you know I [36:50] wish it had is like you know sort of [36:52] like um remember like a MySpace top [36:54] eight like I want that for business like [36:56] who who you have trenches with who do [36:58] you talk to regularly like who who are [37:00] mutuals h how do you know them? How'd [37:01] you meet them? Um you know what do they [37:03] think about you? There's more as sort of [37:05] as you said it treats connections all [37:07] all the same. Um and you don't really [37:10] get context on the relationship or [37:11] context on you know people's reputation. [37:13] It's like LinkedIn is about personal [37:14] identity. We hope that cosign is about [37:16] you or I'm sorry professional identity. [37:17] We hope cosign is about professional [37:19] reputation. You know, LinkedIn is about [37:21] they say it's like people you know. [37:23] Twitter is like uh people who are [37:25] interested in knowing you and we hope [37:26] that cosign delivers both in a more [37:29] granular way. And at the same time, you [37:31] to give props to LinkedIn, it's like it [37:33] started 20 plus years ago. [37:34] >> Yeah. [37:35] >> And no one's even come close to [37:37] competing with it in any capacity. I [37:39] remember Charles Hudson uh you know [37:41] great great investor like you know 12 [37:43] years ago was a blog post like where are [37:44] the LinkedIn competitors it's time to [37:46] disrupt like I don't like the product [37:47] for XYZ reasons it doesn't do this and [37:50] we've seen tons of them you know we [37:51] backed them you know at the firm [37:52] individually and uh you know people try [37:54] crypto like all sorts of like different [37:56] incentives to to break LinkedIn and it [37:59] it's just fascinating I you know I was [38:00] joking when I said everyone hates it but [38:01] people do have like they use it [38:04] obsessively and yet they have like low [38:06] NPS on it or or thing and they know that [38:09] a better version can exist and yet you [38:12] know it's been impossible to to disrupt [38:14] >> but but I think the real thing with [38:16] LinkedIn is actually think the NPS is [38:17] really high for the people who aren't [38:20] actively in the flow and in the [38:22] searching and in the community. [38:23] LinkedIn's always served two audiences. [38:25] One is I call it the people who want to [38:27] find people and the other one is people [38:29] who want to maybe be found at some point [38:31] in their lives. And when you think about [38:32] it all of us would love to be found for [38:35] the right opportunity that comes to us. [38:37] We aren't no not a lot of us might be [38:39] happy. We used to talk about the like [38:41] mildly happily employed because there's [38:43] a lot of people who like they like what [38:45] they're doing. They're not in any urgent [38:47] need to go do something else. But if the [38:49] right thing popped, whether it was from, [38:51] you know, someone you know or someone [38:52] you don't know, but it matched all your [38:54] interests, you might be like, hm, should [38:57] I actually think about that? And I might [38:58] consider it. It still may not be right [38:59] for you in life at the time. But for [39:01] those mildly happily employed people, [39:03] they aren't constantly, you know, if [39:04] they're an engineer, they're not [39:06] checking LinkedIn every day for like new [39:07] people or new things. If if you're in [39:09] sales, if you're in BD, if you're in [39:11] venture and investor, you're using it [39:13] all the time. But but that was what was [39:15] really special. And and I think moving [39:18] tens and tens and tens or hundreds of [39:20] millions, I think they're over a billion [39:21] people now, of people who are happily [39:24] employed, who are maybe open to the [39:26] right opportunity, like that's the kind [39:28] of network effect that really sticks. [39:30] and and you think a lot about what is a [39:32] 20-year network effect like look at [39:34] LinkedIn and as you're building and and [39:36] I say this a lot to to new companies as [39:38] you're thinking about something new how [39:40] do you think about that kind of network [39:41] effect that could be so powerful for 20 [39:43] years and even if some people wish there [39:45] was a better product you can't move the [39:47] masses over [39:48] >> yeah we have looked at and invested in [39:50] some attempted LinkedIn killers and I [39:52] think what they've done a remarkable job [39:54] with is the switching costs are just [39:56] very high both because they are famously [39:59] good at bot detection [39:59] >> [laughter] [40:00] >> So, you can't really scrape it in real [40:02] time. There's whole businesses that are [40:04] built off of scraping LinkedIn and [40:06] selling that data. Um, but also because [40:08] it's just such a tight and dense network [40:11] that like ultimately if it's something [40:13] as important as your career, you're [40:14] going to want to be on the platform [40:15] where everyone else is. That being said, [40:17] I think there's a massive opportunity [40:19] for a new platform when there is new [40:21] information or new value to be kind of [40:24] gleaned or shared which is like very [40:26] much highquality endorsements and high [40:28] conviction people [40:29] >> generational shifts uh and also uh niche [40:32] the the hyper niche that needs it own [40:35] LinkedIn. For example, GitHub is [40:36] arguably a better LinkedIn for engineers [40:38] than than LinkedIn itself. And I'd say [40:39] is there an opportunity to say like what [40:41] is the emergent niche here? Is it the um [40:44] the individual who's trying to build [40:46] their reputation as a talent scout? [40:47] They're trying to be a founder who wants [40:49] to break through. could this is there a [40:50] world for a better LinkedIn that [40:51] represents their I yeah their taste in [40:54] people the [40:56] >> and what gets me excited about cosign [40:57] isn't that it has to even be a better [40:58] LinkedIn it's that it really pulls [41:01] together all this information that's [41:03] already latently happening in our [41:04] industry and and look Silicon Valley and [41:06] the tech industry is so special for this [41:09] kind of pay it forward for this belief [41:11] in somebody the stuff we talked at the [41:12] very beginning the belief in this person [41:15] who may not have had a previously strong [41:18] reputation but has an insight an idea [41:20] and that alone can gather a bunch of [41:22] people and a bunch of momentum and [41:24] change the world and and that's what [41:26] makes this so special and that's what I [41:28] think you know these platforms like [41:29] LinkedIn you know serving so many people [41:31] now across so many industries and so [41:33] many domains and I think there's room [41:35] for something that's really special for [41:37] the the community that we have and that [41:39] you know combines the things that are [41:40] happening ephemerally with the things [41:42] that are happening professionally with [41:43] the depth that that is going to be great [41:45] and if cosign works great and it still [41:47] only serves the tech That's a huge win. [41:50] >> Oh, 100%. [41:50] >> And and and if it then other communities [41:53] and other industries figure out they [41:54] should figure out how to take part in [41:56] this too and and bring some of that [41:58] Silicon Valley energy to more places, [42:00] that's even better. But the the pay it [42:02] forward and the every great idea can [42:04] just come from somebody who wasn't [42:06] previously known but truly has a unique [42:08] insight. Like that's what I hope this [42:10] captures too. E Eric uh from Benchmark [42:13] had this comment in the Patrick Ocean [42:15] podcast where Patrick was asking him [42:16] sort of like you know what do you think [42:18] about every category and he's like it's [42:20] all gonna work like everything is going [42:22] to be be huge not every company but just [42:23] as an index like every c I think we've [42:25] come to similar conclusion um and I [42:28] think and I think similarly like when [42:30] people say like uh you know do does the [42:32] world need another podcast like there's [42:33] too many podcasts oh and yet another [42:35] great podcast comes in and it doesn't [42:36] kill the other it just all works uh [42:38] everything that's great works and and [42:39] similarly like you know we just launched [42:41] school like Stanford's going to be fine [42:43] you know our going to be fine and our [42:45] thing's going to be great and then [42:46] similarly like yeah LinkedIn's going to [42:48] be great like it's it's going to keep [42:49] doing it thing and you know we're [42:51] inspired by them in so many ways we're [42:52] also inspired by Angelist like 10 years [42:54] ago when they owned the investor graph [42:57] and really really it was for the startup [42:59] community and it was really focused [43:00] around you know companies investors and [43:02] talent [43:03] >> and you know they made a series of other [43:05] bets that then sort of um made them [43:08] focus on other parts of that or [43:09] deprecate the the the the you the free [43:11] public social network that used to exist [43:12] and we think it's an opportunity to [43:13] bring back and then we're also inspire [43:15] in terms of like how we're going to do [43:16] this. We're also inspired by Wikipedia. [43:19] Um, one in terms of, you know, people [43:21] want Wikipedia profiles, they're [43:23] aspirational, but two, in that we're [43:24] creating profiles for people. Um, so [43:26] that that's where we're starting and we [43:28] think people are going to be excited [43:28] about it because we're bragging for them [43:30] on their behalf in an exclusive, you [43:32] know, way. And, you know, if they don't [43:34] want it, we we'll take it off, but [43:35] you're going to see your profile. it's [43:36] going to be flushed out with all all you [43:37] know people giving you compliments and [43:38] also people expressing their interest in [43:40] funding you or working with you or like [43:42] even before you go to the profile like [43:43] you know before you know and so that's [43:46] one of the ways in which we're [43:47] differentiating and only you know [43:48] because of AI can we generate all of [43:51] this information on people and like [43:53] recreate you know a professional graph [43:55] um but if we didn't go that third party [43:57] route then we would have to get people [43:58] to sign it's cold start problem it's the [43:59] place where people go to update their [44:01] you know professors it's the most [44:02] reliable often people say like what's [44:04] the daily behavior well we're, you know, [44:06] brute forcing it by by covering every [44:09] sort of, you know, transition. So, we [44:11] will also have reliable profiles with [44:13] uh, you know, more reputation benefits [44:15] to it. [44:16] >> Totally. I feel like the best I'm [44:18] curious if you agree, but for consumer [44:19] products, like the easiest way to [44:21] acquire users is to tell them that [44:23] someone else is saying something about [44:24] them on the product, [laughter] [44:26] especially if it's something it's the [44:27] Nikita beard like your friends just [44:30] answered this question about you [44:31] strategy, which worked so well three [44:33] times. Um and and it wasn't possible [44:36] before AI for for a network I think as [44:38] big as Silicon Valley and now it is [44:40] which is awesome. [44:40] >> Someone calling me today the Nikita [44:41] Bureau founder communities. [laughter] [44:43] >> That's a real compliment. [44:44] >> That's that's a true compliment. [44:47] >> Yeah. So it's worth pointing out um [44:49] everything that is being pulled in is [44:50] coming from public sources. So it's what [44:51] people have posted on their uh you know [44:54] variety of different places. Um it's [44:55] being aggregated. Uh there's a lot of [44:57] like the funding announcements. They're [44:58] the ones that being announced. Um so [45:00] there's a lot of really really high [45:02] quality information that is spread [45:03] across the publicly available internet [45:04] and it's is uniquely uh compiled and [45:07] structured we think in one place and [45:08] Eric talks about the angelist example. I [45:10] recall very early in my career angelist [45:12] was probably the first it was like the [45:13] wedge that I took into the Silicon [45:15] Valley at all. was the it was the um [45:17] ability to see which companies and which [45:19] geographies which people at those [45:20] companies which um you know for for a uh [45:23] for a long time it it felt like it could [45:25] be the LinkedIn for this sort of little [45:27] niche of people who were ambitious in [45:29] technology and then as Eric said it sort [45:30] of it moved away from that for a number [45:32] of reasons but that's been missing a [45:33] place that you can go as a founder who's [45:35] like uh which angel investors do I want [45:37] to get to know over the next 6 months [45:39] before my round it's a public utility [45:41] something it's worth pointing out here [45:42] you know we we're putting this out into [45:44] the world it's it's like a um a thing [45:46] we'd like to build with people. It's not [45:48] something that's being presented as a a [45:50] fully finished perfect products. You [45:52] know, we it's an invitation. It's an [45:53] invitation to everybody who wants to [45:54] come in and you know edit and you know [45:57] submit [45:59] an update to to their own or someone [46:00] else's profile. Uh it's a real ask like [46:03] let's build this thing together for for [46:04] everyone. [46:05] >> Yeah. and and submit things to people's [46:07] profiles, cosign the things you endorse [46:09] on people, and then even if somebody has [46:11] feedback, cosign that feedback so that [46:13] it gets back to [laughter] David and the [46:15] team in order to to keep improving [46:18] things. [46:18] >> Yeah. And so sort of to end on that sort [46:20] of higher level note, uh a lot of the [46:23] industry but also other industries sort [46:24] of um run on cosigns in you know uh Paul [46:27] Graham co-signed Sam Alman famously uh I [46:30] feel like 2009 or something he said like [46:32] the five best entrepreneurs like you [46:34] Steve Jobs you know three other people [46:35] and Sam Alman and at the time Sam Alman [46:37] hadn't like had a huge exit yet but that [46:39] really just like gave him such a [46:41] reputational boost him running YC etc [46:44] you know is critical to him then [46:45] building going on to build one of the [46:46] most important companies of all and and [46:49] so many you know Ben Haritson Ali from [46:51] data bricks I mean so many examples of [46:53] people who got their cosign from someone [46:56] older and more senior in the industry [46:57] and were able to rise up because of it [46:59] >> Ryan Hoover who was founder of product [47:01] hunt co-signed Eric and brought Eric on [47:03] to become one of the leaders of building [47:05] the community for everything and it was [47:06] such a great endorsement of you early in [47:09] your career and your potential I mean [47:10] look at what you've become since it's [47:11] amazing [47:12] >> I'm grateful for him and um it it's it's [47:15] so fascinating because I mean one of the [47:17] things that makes Silicon Valley so [47:18] special is how much of a meritocracy it [47:20] is and how you don't have to like you [47:22] know necessarily just you know pay your [47:24] dues for 20 years before you take it [47:26] seriously like one of the most important [47:27] intellectuals right now is Doresh he's [47:30] 25 years old he's never had a job as far [47:32] as I'm aware he just you know came to [47:33] San Francisco started podcasting and you [47:36] know someone was if I don't know if it's [47:38] Patrick Olson or Marson also went on [47:40] like a number of people by going on his [47:41] podcast early cosigned him and then said [47:43] hey this is a place where then Zuck and [47:45] others can go on you know the one of the [47:47] most important founders is Michael Troll [47:49] you know a young guy in his you know mid [47:51] to late 20s etc. one of the most [47:53] important uh you know people labelled [47:55] Ashen Brener is like I think like 24 25 [47:58] as well and so it just like if you're [48:00] talented if you're smart in this [48:02] industry um someone will discover you [48:06] cosign you and make your career and so [48:08] we hope that happens in our [48:09] >> well we're talking about social capital [48:10] and positive some mindsets I think it's [48:11] a really important um there's a critique [48:13] of of this I I wanted to to talk about [48:15] where someone might rightfully say well [48:17] why would I share this this cosign this [48:19] is my alpha this is like I know that [48:21] person's good like why would I tell [48:22] you're you're good. And I think actually [48:24] something um if if you take a very [48:25] long-term view at this, you can say um [48:27] sure if that person's raising a preede [48:29] round right now and you want to do it, [48:30] don't cosign them right now, but you the [48:32] person you want to to cosign is the one [48:34] that's raising a preede round two years [48:36] from now, 5 years from now. And when [48:38] they do, they remember you. They say, [48:40] "Yeah, the reason I'm able to raise this [48:41] free seed round is Olivia cosigned me [48:43] two years ago." And in doing so, she [48:45] helped me meet my co-founder and she [48:47] helped me meet the five people who were [48:49] my early believers in my product. And [48:50] that's why I'm here today. And of course [48:52] I'm going to call her first my preceding [48:53] ground. And if you turn it around that [48:55] way and you take a very long-term [48:56] mindset like it's such a such a [48:57] long-term [48:58] >> 100% the benefits just acrew to you in [49:01] the sense of people always remember like [49:03] by order of magnitude more the people [49:04] who believed in them first right you [49:06] know Paul Brian Chesy when he uh you [49:09] know goes public with Airbnb who does he [49:11] write in the in the blog post he writes [49:13] YC you know Paul his first investor he's [49:15] had you know so much investment all [49:17] throughout the book so much more money [49:18] so much more time uh into this company [49:20] but you remember the first person the so [49:23] and so if you build if you are the first [49:25] person especially publicly cuz that's [49:26] where the value is uh because they will [49:28] remember you for it and and then more [49:30] people will come to you. [49:32] >> Yeah. So let's wrap on that. Thank you [49:35] so much. Thank you. [49:36] >> Thank you.