Canales actualizados recientemente

Elige uno para ver solo ese canalVer todos los canales →

Resultados

25 artículos
Ordenar
1:59:52EN
Zhang Xiaojun Podcast· 21 oct 2026Pódcast · 2 vocesDavid Qian on Succession, Robotics, and Finding His Own AmbitionInvitado: David Qian (Qian Cheng) · Presenta: 张小珺Ecovacs CEO David Qian reflects on inheriting a founder-built company, learning to lead through mistakes, and choosing a patient path in robotics. His account treats succession as an ongoing effort to build the company for a new era, not a finish line.Empresa y estrategiaLeadershipCompany Building+2
2:51:18EN
Zhang Xiaojun Podcast· 15 oct 2026Pódcast · 2 vocesLindon Gao on Building Dyna Robotics and Finding a Path Through the FogInvitado: Lindon Gao · Presenta: 张小珺Dyna Robotics CEO Lindon Gao explains why his company builds intelligence through real workflows, how hardware and research must evolve together, and how a founder can find clarity by letting go of the need to be right.RoboticsProduct StrategyModelos y entrenamiento+2
2:34:18EN
Zhang Xiaojun Podcast· 3 sept 2026Pódcast · 2 vocesZeng Ming on AI’s New Era: Why Great Companies and New Organizations EmergeInvitado: Zeng Ming · Presenta: 张小珺Zeng Ming argues that AI is changing the foundations of business: models will become infrastructure, while new Agent-native applications and organizations create the next wave of value. His advice to founders is to embrace uncertainty, rethink strategy continuously, and build around meaningful work rather than inherited structures.AI StrategyEmpresa y estrategiaAgentes+2
2:34:18ZH
Zhang Xiaojun Podcast· 3 sept 2026Pódcast · 2 voces曾鸣:AI 产业化的三阶段与公司组织的消亡Invitado: 曾鸣 · Presenta: 张小珺战略学者曾鸣以互联网、工业革命和电气化的历史周期为参照,提出 AI 正从基础设施走向 Agent 应用,并最终进入原生应用阶段。他认为,旧公司的科层制难以穿越范式转型,下一代组织将围绕任务、创造力和人机协同重构。AI StrategyStrategyAgentes+2
3:36:13EN
Zhang Xiaojun Podcast· 13 ago 2026Pódcast · 2 vocesNVIDIA’s Ming-Yu Liu on Cosmos, Physical AI, and Creating New MarketsInvitado: Ming-Yu Liu · Presenta: 张小珺NVIDIA research VP Ming-Yu Liu explains how a decade of generative-model research led to Cosmos 3, a world foundation model for Physical AI. He also shares what he learned from Jensen Huang about long-term bets, teamwork, and building markets instead of fighting over existing ones.World modelsRoboticsAI research+2
2:59:03EN
Zhang Xiaojun Podcast· 28 jul 2026Pódcast · 2 vocesYou Kaichao on How vLLM Grew from Open-Source Project to AI Infrastructure CompanyInvitado: You Kaichao · Presenta: 张小珺Inferact co-founder You Kaichao traces vLLM’s path from Berkeley research to a company, then explains why model, inference, hardware and agent design increasingly have to move together.Company BuildingAI startups & small teamsAI research+2
3:46:12EN
Zhang Xiaojun Podcast· 16 jul 2026Pódcast · 2 vocesKay Ke on Physical Intelligence, Robot Learning, and the Future of RoboticsInvitado: Kay Ke · Presenta: 张小珺Physical Intelligence researcher Kay Ke traces the field's research lineages and explains Pi's progression from π₀ to π*₀.₆. She also reflects on education, fiction, ambition, and what it means to build robots that work beyond a demo.RoboticsReinforcement LearningAI research+2
46:26ZH
Zhang Xiaojun Podcast· 9 dic 2025Pódcast · 2 voces朱啸虎:三年内看不到泡沫,共识太集中就错开十五度Invitado: 朱啸虎 · Presenta: 张小珺2025 年底,金沙江创投朱啸虎第三次坐到张小珺对面:AGI 不谈了,入口之争开始了。他不投人形机器人,把支票开给洗船和按摩这类「干活」的公司。Venture CapitalAI StrategyOpenAI+2
2:06:16ZH
Zhang Xiaojun Podcast· 28 nov 2025Pódcast · 2 voces谭捷:机器人通用智能的瓶颈在数据,通用形态仍需时间Invitado: 谭捷 · Presenta: 张小珺Google DeepMind 机器人研究科学家谭捷回顾了从计算机图形学到机器人、从强化学习到 Gemini Robotics 的研究路径。他认为机器人基础模型尚未成为独立学科,数据规模与覆盖度、跨本体迁移、实时推理和安全仍是走向真实应用的关键。RoboticsFoundation ModelsWorld models+2
1:48:46ZH
Zhang Xiaojun Podcast· 18 nov 2025Pódcast · 2 voces刘先明:拆掉语言,让物理 AI 回到数据与动作Invitado: 刘先明 · Presenta: 张小珺小鹏自动驾驶中心负责人刘先明解释,为什么 VLA 训练中间的语言监督会成为效率瓶颈,以及他如何推动小鹏从自动驾驶团队转向物理 AI。访谈也谈到规模化数据、组织变革、量产安全与自动驾驶的下一阶段。Autonomous DrivingAI StrategyRobotics+2
1:09:11ZH
Zhang Xiaojun Podcast· 18 ago 2025Pódcast · 2 voces广密谈大模型分化、AI 产品与创业窗口Invitado: 广密 · Presenta: 张小珺广密复盘模型公司的战略分化、ChatGPT 的消费品牌壁垒,以及 Claude Code 等产品带来的新机会。他认为 AI 创业窗口正在缩短,但产品体验、团队判断和非共识创新仍能打开空间。AI StrategyProduct StrategyProducto y UX+2
1:32:16ZH
Zhang Xiaojun Podcast· 20 jun 2025Pódcast · 2 voces王忻谈奔驰中国研发:在转型中重划豪华与安全Invitado: 王忻 · Presenta: 张小珺奔驰中国研发负责人王忻回顾中国汽车产业二十年的技术周期,也讲述中国团队如何从测试支持走向产品研发。面对智能化竞争,他认为速度、数据和本地化必须建立在安全与用户体验之上。Autonomous DrivingSelf-Driving CarsProduct Strategy+2
Diapositivas34:07EN
Zhang Xiaojun Podcast· 26 abr 2025Pódcast · 3 vocesOla Källenius on Mercedes-Benz’s Next Transformation: EVs, AI, and Modern LuxuryInvitado: Ola Källenius · Presenta: 张小珺Mercedes-Benz CEO Ola Källenius describes a century-scale shift driven by zero-emission powertrains, software, and AI. He explains why the company is expanding electric models while protecting the craftsmanship, safety, and desirability that define its luxury position.Autonomous DrivingProduct StrategyEmpresa y estrategia+2
2:29:41ZH
Zhang Xiaojun Podcast· 7 abr 2025Pódcast · 4 voces陈建宇:机器人基座模型如何走向通用 VLAInvitado: 陈建宇 · Presenta: 张小珺从 SayCan、RT-2 到 Diffusion Policy,清华助理教授、星动纪元创始人陈建宇梳理机器人模型从借用语言模型到端到端 VLA 的路线。关键难题仍是多样化数据、跨本体泛化和高频动作控制。RoboticsModelos y entrenamientoMultimodal AI+2
2:36:13ZH
Zhang Xiaojun Podcast· 24 feb 2025Pódcast · 2 voces松琳解读 DeepSeek、Kimi 与 MiniMax 的注意力新论文Invitado: Songlin Yang(松琳) · Presenta: 张小珺从动态稀疏注意力到线性注意力混合架构:三条长上下文路线如何兼顾训练、推理与硬件效率Transformers & attentionGPU programming & kernelsModelos y entrenamiento+2
53:07ZH
Zhang Xiaojun Podcast· 17 feb 2025Pódcast · 2 voces张亚勤:信息智能、物理智能与生物智能的 AGI 路线图Invitado: 张亚勤 · Presenta: 张小珺张亚勤把 AGI 的发展分为信息、物理和生物三类智能:信息智能可能在五年内接近人类,通用机器人约需十年,生物智能则可能在二十年后扩展人的记忆与健康。他也讨论大模型如何为自动驾驶和机器人补上数据、常识与行动能力。AGIRoboticsAI research+2
2:49:35ZH
Zhang Xiaojun Podcast· 4 feb 2025Pódcast · 2 voces潘嘉怡解读 R1、o1 与 Kimi K1.5:强化学习如何教模型学会推理Invitado: 潘嘉怡 · Presenta: 张小珺伯克利博士生潘嘉怡带读 DeepSeek、OpenAI 与 Kimi 的推理模型报告,拆解长思维链、GRPO、奖励设计与蒸馏。R1-Zero 的“顿悟时刻”展示了推理策略如何在强化学习中涌现。Reinforcement LearningModelos y entrenamientoAI research+2
1:27:13ZH
Zhang Xiaojun Podcast· 29 dic 2024Pódcast · 2 voces李想:汽车将从交通工具进化为空间机器人Invitado: 李想 · Presenta: 张小珺理想汽车 CEO 李想谈公司为何从造车走向人工智能:基座模型是 AI 时代的操作系统,汽车是进入物理世界的重要载体。对他而言,技术路线最终要落实为更好的用户体验、能力与长期价值。AI StrategyProduct StrategyAutonomous Driving+2
1:27:02ZH
Zhang Xiaojun Podcast· 5 sept 2024Pódcast · 3 voces广密:Self-play RL 可能开启 AGI 的新 Scaling LawInvitado: 广密 · Presenta: 张小珺广密认为,纯靠语言预训练扩展模型的边际收益可能正在减弱,self-play 强化学习是更值得押注的新路径。它能否泛化到代码、视频、机器人和更广的领域,将决定下一阶段的 AGI 竞赛。Reinforcement LearningAGIScaling Laws+2
1:13:06ZH
Zhang Xiaojun Podcast· 17 ago 2024Pódcast · 2 voces施宏俊:把芒格带到中国,也借芒格重启人生Invitado: 施宏俊 · Presenta: 张小珺《穷查理宝典》在中国售出超过 115 万册,背后是出版人施宏俊对内容、读者和长期主义的判断。他回顾出版黄金年代、行业渠道变迁,以及芒格如何影响他离开管理岗位,创办芒格书院。InvestingHistoryCareer Growth+2
1:08:29ZH
Zhang Xiaojun Podcast· 17 ago 2024Pódcast · 2 vocesColin:企业内斗与内耗,如何从竞争走向消耗Invitado: Colin · Presenta: 张小珺从战略、制度、组织文化和个人处境出发,Colin 与张小珺讨论竞争何时有益,以及员工如何识别和应对内耗。ManagementCompany CultureLeadership+2
1:33:07ZH
Zhang Xiaojun Podcast· 17 ago 2024Pódcast · 2 voces广密复盘全球大模型这一年:Scaling Law、AGI 豪赌与中美格局Invitado: 李广密 · Presenta: 张小珺拾象创始人李广密从硅谷一线观察出发,回顾 OpenAI、Anthropic、Google 与中国模型公司的竞争。他认为,模型能力与成本共同决定新一轮应用爆发,而巨额投入押注的是智能能否持续扩展以及 AI 对经济和产业的长期重构。Modelos y entrenamientoScaling LawsAI research+2
1:14:56ZH
Zhang Xiaojun Podcast· 17 ago 2024Pódcast · 2 voces王亚军解读黄仁勋的组织系统:用分布式决策支撑创新Invitado: 王亚军 · Presenta: 张小珺前龙湖集团 CSO 王亚军从战略、产品与组织的关系出发,拆解英伟达不裁员、六十名高管直报和少做计划等做法。关键不在照搬形式,而在于建立真实、信任和横向协作,让更多人掌握决策所需的信息。ManagementEmpresa y estrategiaLeadership+2
49:41ZH
Zhang Xiaojun Podcast· 17 ago 2024Pódcast · 3 voces傅盛:大模型是生产力革命,机器人迎来新的窗口Invitado: 傅盛 · Presenta: 张小珺傅盛认为,大模型的核心是逻辑与推理能力,它带来的变化更像蒸汽机,而不是一次普通的互联网迭代。从平民化模型、服务机器人到创业者如何面对低谷,他也回看了自己的判断与心境。AI StrategyModelos y entrenamientoRobotics+2
Cargar másMostrando 24 de 25