筛选结果
29 篇排序
1:09:44EN10,000 GB300s for Four Weeks: Reflection’s Open-Model Ledger主持 Sarah Guo, Elad GilMisha Laskin tells No Priors how Reflection trained Beam, why reinforcement learning outspent pre-training, and why the West needs its own open weights.
4:18:22EN → 简体中文每个周期都重要:解码几乎全网视频的志愿者们嘉宾 FFmpeg · 主持 Lex FridmanLex Fridman 与 Jean-Baptiste Kempf、Kieran Kunhya 谈 FFmpeg、VLC、手写汇编,以及 YouTube、Netflix 和火星车背后那群没有报酬的人。
44:46ENA Single Training Environment Can Cost Millions. Prime Intellect Wants a GitHub for RLWill Brown and Johannes Hagemann argue that the eval suite you already use to test agents is one step away from a training set.
43:09ENA $100 Arm and a $300 Robot: Hugging Face Wants an App Store for Physical AIThomas Wolf tells Sequoia's Training Data why LeRobot, cheap hardware, and open datasets are Hugging Face's bet that software developers can become roboticists.
6:08:48EN → 简体中文没人认真相信 Cookie 横幅有用嘉宾 DHH · 主持 Lex FridmanLex Fridman 与 David Heinemeier Hansson 聊了六个小时,话题涵盖 Ruby、Rails、AI、离开云、苹果的 App Store,以及小公司为什么依然可以拒绝别人的钱。
5:20:08ENThePrimeagen Would Rather Redesign the Room Than Trust Willpower嘉宾 ThePrimeagen · 主持 Lex FridmanLex Fridman talks with Michael Paulson about linked lists, Netflix, Neovim, ADHD, addiction, AI, and the night he says God gave him a conscience.
42:08ENMeta Trained a 405B Teacher on 16,000 GPUs — Then Told the World to Distill ItTwo days after Llama 3.1 405B shipped, Meta's generative-AI product lead walks Sequoia through why the most expensive model in the family is free, and why he thinks the frontier itself is heading toward commodity.
32:57ENA 7-billion-parameter model on a MacBook: Mistral's CEO reruns the cost of AI主持 Sarah Guo, Elad GilArthur Mensch tells No Priors why Chinchilla made small models valuable, why open-sourcing Mistral 7B is not a bioweapon risk, and why guardrails belong on the application.
1:01:09ENHugging Face Raised $235M. Community Still Is Not a Moat主持 Nathan LabenzNathan Labenz and Erik Torenberg unpack Hugging Face's $4.5 billion Series D, and why a thriving open-source hub may still struggle to become a locked-in business.
3:08:46EN → 简体中文谁掌握关机开关?George Hotz 谈 tinygrad、开源 AI 与上帝嘉宾 George Hotz · 主持 Lex Fridman这是他第三次做客 Lex Fridman 的播客。comma.ai 和 tiny corp 的创始人主张,算力、智能和关机开关应当留在普通人家里,而不是掌握在少数可信之人手中。
3:15:07EN → 简体中文写作不再像读研究生,维基百科才跑了起来嘉宾 Jimmy Wales · 主持 Lex Fridman吉米·威尔士在与莱克斯·弗里德曼的对谈中,讲了 Nupedia 七阶段审稿为何走向终结,维基百科为何至今不挂广告、也不向政府低头,以及大语言模型到底该在百科全书里做什么。
2:07:04EN → 简体中文核糖体早就解决了数字化制造嘉宾 Neil Gershenfeld · 主持 Lex FridmanNeil Gershenfeld 与 Lex Fridman 对话,谈图灵(Turing)为什么要把读写头和纸带分开,fab lab 为何是制造领域的“个人电脑时刻”,以及用大约二十种零件培育一个文明需要些什么。
37:26ENWithout Open Source, Machine Learning Would Be Decades Behind主持 Sarah Guo, Elad GilHugging Face CEO Clem Delangue on the Tamagotchi pivot, why the company never hired a community manager, and why he does not believe one model will rule them all.
45:03ENEmad Mostaque’s bet: open small models, not one giant brain主持 Sarah Guo, Elad GilStability AI’s founder tells No Priors why foundation models should be public infrastructure, why small models can beat giant ones, and what that means for media, medicine, and AGI.
3:15:51EN → 简体中文Python 是写给两类读者的一份菜谱嘉宾 Guido van Rossum · 主持 Lex Fridman吉多·范罗苏姆(Guido van Rossum)谈 CPython 3.11 的提速、缩进、GIL,以及为什么 4.0 对扩展模块的伤害会大过语法本身。
2:46:40EN → 简体中文能装进一辆面包车的人,造出了能装进脑子的工具嘉宾 Peter Wang · 主持 Lex FridmanAnaconda 的 Peter Wang 与 Lex Fridman 谈 Python、赛博系统,以及为什么分享软件能创造价值,而稀缺经济学至今仍无法为其定价。
3:05:24ENTravis Oliphant wrote NumPy in an empty semester — and still wants open source to pay嘉宾 Travis Oliphant · 主持 Lex FridmanThe creator of NumPy, SciPy, and Anaconda tells Lex Fridman how array thinking, a priced book, and the packaging wars shaped scientific Python.
2:21:27EN → 简体中文IBM 手册写得太差,他索性成了计算机科学家嘉宾 Donald Knuth · 主持 Lex FridmanDonald Knuth 与 Lex Fridman 对谈:他在 IBM 650 上写的第一批程序、文学化编程(literate programming)、Knuth-Morris-Pratt 算法,以及他证明过的最难的一条定理。
2:54:52EN → 简体中文精彩的证明,仍藏在页边的空白里主持 Lex Fridman费马图书馆的两位创始人 Luís 与 João Batalha 与 Lex Fridman 谈论论文背后的隐藏故事、一种把自己的作品买了两次的出版体系,以及如何死磕一篇论文,直到它终于豁然开朗。
3:03:38EN → 简体中文只花十天写成的语言,至今仍在驱动网页嘉宾 Brendan Eich · 主持 Lex FridmanBrendan Eich 在与 Lex Fridman 的对谈中讲述,JavaScript、Firefox、Mozilla 和 Brave 是怎样从仓促的赌注、开源(open source)逃生舱和意外的追踪机制中诞生的。
1:51:01ENJames Gosling Built Java to Make Pointer Bugs Impossible. Chrome Still Has Them.嘉宾 James Gosling · 主持 Lex FridmanIn a 2020 conversation with Lex Fridman, Java’s designer traces a line from a neglected PDP-8 to Emacs, the JVM, and a thirty-year war on memory errors.
1:43:10ENKen Thompson's Wife Left for Three Weeks. He Wrote Unix.嘉宾 Brian Kernighan · 主持 Lex FridmanBrian Kernighan walks Lex Fridman through Bell Labs, C, AWK, AMPL, and a lifetime of tools that millions of people still run — including all of his mistakes.
1:47:13ENWhen You Hit the Brakes, a Computer Is in the Way嘉宾 Bjarne Stroustrup · 主持 Lex FridmanBjarne Stroustrup on why C++ still sits between the pedal and the pads, and why an abstraction that costs extra has no place in the language.
1:10:58ENThe Hadoop Lesson That Made Google Give TensorFlow Away嘉宾 Rajat Monga · 主持 Lex FridmanIn November 2015, Google put its core internal machine-learning system on GitHub. Rajat Monga explains why a company of that size would hand over the hard-to-copy part.
加载更多已显示 24 篇,共 29 篇









