Derniers articles
3 792 articlesTrier
50:35ENCopilot Already Writes Half the Code. GitHub's CEO Says Agents Still Have Four Gates to ClearAnime : Sarah Guo, Elad GilThomas Dohmke on agent mode, why median developers are not going anywhere, and how AI is repricing the tools that write software.
9:02ENDario Amodei’s 2025 AI Coding Forecast and Matthew Berman’s ResponseMatthew Berman, Dario Amodei, Mark Zuckerberg · Forward FutureBerman considers a rapid forecast for AI-written code, the rise of coding agents, and speculative longer-term changes to software work.
1:12:13ENNiklas Östberg on Delivery Hero’s Growth, Acquisitions and the Cost of Moving Too FastInvité : Niklas Östberg · Anime : Harry StebbingsThe Delivery Hero founder reflects on pandemic-era financing, a costly Gorillas investment and how to distinguish sustainable delivery growth from subsidized volume.
59:46ENIn Young Cho on Trading Research and Finding Signal in Financial DataYaron Minsky, In Young Cho · Jane StreetIn Young Cho and Ron Minsky discuss the blurred lines between trading, research, and engineering, and why machine learning in markets is difficult.
1:30:57ENAnt's $320 Billion IPO Was Stopped at the Eleventh Hour: Angela Zhang on China's Tech TightropeAnime : Nathan LabenzA USC law professor walks through the dynamic pyramid that governs Chinese tech, from Didi's listing to DeepSeek's price war, and why she thinks the United States is starting to look more like China.
27:32ENOpenAI’s Responses API and Agents SDK: A Launch WalkthroughMatthew Berman · Forward FutureMatthew Berman reacts to a launch demonstration of web and file search, computer use, agent handoffs and tracing.
54:10ENHarvey Built Legal AI for the Work the Next Model Still Cannot DoCEO Winston Weinberg on citations, process data, and the unscalable tactics that won over century-old law firms.
1:33:55ENJake Saper on Zoom, AI and the Hard Decision of When to SellInvité : Jake Saper · Anime : Harry StebbingsThe Emergence Capital partner explains collective diligence, fund distributions and the trade-off between liquidity and holding winners.
1:56:38ENo3 took ARC from 20% to 87%. Nathan Labenz sketches early superintelligenceAnime : Nathan LabenzOn Consistently Candid, the Cognitive Revolution host argues that math and coding will be superhuman by 2026, and that a reasoning core plus specialist “intuitive physics” may already be enough.
54:12ENJulian Teixeira on Hiring Sales Teams and Building a Repeatable ProcessInvité : Julian Teixeira · Anime : Harry StebbingsThe 1Password CRO discusses early hires, team goals, outbound prospecting, compensation and inspecting deals as a company scales.
12:30ENQwQ-32B: A Smaller Reasoning Model and Its TradeoffsMatthew Berman · Forward FutureMatthew Berman examines Qwen’s reinforcement-learning recipe, tests a coding task, and compares benchmark scores and hardware requirements with DeepSeek-R1.
12:10ENMercury Coder and Diffusion Language Models: Speed, Demos and LimitsMatthew Berman · Forward FutureMatthew Berman explores Inception’s 2025 coding model, comparing iterative text refinement with autoregressive generation and examining speed claims.
1:31:35ENThirty Years of Research, Collapsed into One SentenceAnime : Nathan LabenzImmunologist Derya Unutmaz on writing a medical patent in half an hour, treating o1 Pro as a professor-level collaborator, and why he no longer trusts his own expertise until he asks the model.
45:06ENMike Schroepfer: AI’s Bottleneck Is Electricity, and One Tanker of Fusion Fuel Could Power the U.S. for a YearAnime : Jacob EffronMeta’s former CTO tells Redpoint why hyperscaler demand is the climate story, why Llama was an access play, and why tomorrow’s CTO still lives or dies on a priority queue.
36:24ENThe Man Behind Humanity's Last Exam Says Superintelligence Strategy Should Copy NukesAnime : Sarah Guo, Elad GilDan Hendrycks tells No Priors that labs can only do basic anti-terror refusals, while the real risk-management problem is a geopolitical deterrent he calls mutually assured AI malfunction.
21:48ENVibe Coding With Cursor and Windsurf: Berman’s 2025 WorkflowMatthew Berman · Forward FutureMatthew Berman shares project rules, testing habits and version-control practices from early experiments with coding agents.
49:39ENAnton Osika on Lovable’s Launch, Early Retention and Building AI Products in EuropeInvité : Anton Osika · Anime : Harry StebbingsThe Lovable co-founder explains the transition from an open-source experiment to a fast-growing app builder, and discusses hiring, fundraising and the limits of early retention data.
1:04:52ENThe Consumer Playbook That Put Medical AI in 100,000 Exam RoomsOpenEvidence CEO Daniel Nadler explains how a free App Store product, trained only on peer-reviewed literature, reached as much as a quarter of U.S. physicians in a year.
1:02:41ENMike Krieger on AI Products, Model Competition and the Future of CodingInvité : Mike Krieger · Anime : Harry StebbingsAnthropic’s chief product officer explains how he thinks about product design, model differentiation and the limits of current AI tools.
3:22:47ZH肖弘:从 Monica 到 Manus,如何在技术变化中创业Invité : 肖弘 · Anime : 张小珺两次访谈记录肖弘的创业经历、AI 产品判断,以及 Manus 发布前对智能体的设想。
1:08:51ENChip Controls Are Meant to Buy Ten Years, Not Win the AI RaceAnime : Nathan LabenzEconomist Noah Smith tells Nathan Labenz that U.S. export controls are a piece of conventional deterrence against China — not a plan to kneecap DeepSeek.
19:32ENGPT-4.5’s Launch: Conversational Style, Benchmarks and CostMatthew Berman · Forward FutureMatthew Berman examines OpenAI’s February 2025 research preview, its selected demonstrations, reported evaluations and high launch pricing.
52:59ENGeorge Bonaci on Growth Experiments, Hiring and Finding New ChannelsInvité : George Bonaci · Anime : Harry StebbingsRamp’s growth leader discusses balancing speed with rigor, measuring results and building a team that can find opportunities beyond an existing playbook.
2:33:39ENThe Attacker Goes Second: Why Nicholas Carlini Keeps Breaking AI DefensesAnime : Nathan LabenzA Google DeepMind security researcher on simple loss functions, 70% robust accuracy, and why the fix has to live in the system, not the model.
Charger plus24 sur 3 792 affichés









