Résultats
101 articlesTrier
52:30EN
Sequoia Capital· 28 janv. 2025PodcastMove 37 looked like a bug. LLMs still need their AlphaZero momentIoannis Antonoglou, from AlphaGo to Reflection AI, on self-play, reliability, and why compute can still buy intelligence.
52:06ENSequoia Capital· 21 janv. 2025PodcastLinkedIn spent five years on graph neural nets. She wants any company to see value in four weeksKumo co-founder Hema Raghavan argues that the graph was always sitting in your warehouse tables — and that a SQL-like predict clause is how ordinary companies finally get to use it.
1:00:05ENSequoia Capital· 14 janv. 2025PodcastIf Someone Builds a Better Spark Product, We Deserve to LoseIon Stoica on how Databricks turned academic open source into a data-and-AI company — and which problems he thinks will matter more tomorrow than today.
51:37ENSequoia Capital· 10 déc. 2024PodcastForty hours to 28 minutes: the Copilot creator teaching AI to hackOege de Moor tells Sequoia how XBOW’s AI pentest matched a top human on 104 original challenges, and why offensive security is becoming always-on software.
38:48ENSequoia Capital· 3 déc. 2024PodcastWho Wants to Chat With an Expense Report?Ramp CEO Eric Glyman tells Sequoia why the chatbot gold rush missed the point, and why the better analogy for company finance is a Waymo: self-driving money.
1:03:07ENSequoia Capital· 26 nov. 2024PodcastReasoning Hasn't Budged in Two Years. Dust Still Saved a Customer 8,000 HoursDust co-founders Gabriel Hubert and Stanislas Polu on multi-model agents, refusing to buy GPUs before product-market fit, and the jump from calculator tools to stochastic ones.
51:38ENSequoia Capital· 19 nov. 2024PodcastClay Wants to Be a Guitar, Not a Microwave, for Go-to-Market TeamsKareem Amin tells Sequoia’s Alfred Lin why Clay is building a system of action for growth rather than an AI that replaces SDRs.
46:35ENSequoia Capital· 13 nov. 2024PodcastThey trained a real-time video model in 20 hours and skipped the game engineDecart CEO Dean Leitersdorf tells Sequoia how Oasis runs live on H100s, why ChatGPT is a fallen wall rather than a pain point, and why the next trillion-dollar company may be a generated experience.
44:48ENSequoia Capital· 29 oct. 2024PodcastEnterprise Google Failed for Twenty Years. A Search Veteran Finally Made It Work.Glean CEO Arvind Jain on why company search is harder than the public web, why RAG lives or dies on retrieval, and what work looks like when every employee has an AI assistant.
41:42ENSequoia Capital· 22 oct. 2024PodcastNeural Networks Are Still Waiting for Their ThermodynamicsOpenAI researcher Dan Roberts on why physicists keep migrating into AI, and why the field still lacks the equivalent of the ideal gas law.
32:07ENSequoia Capital· 15 oct. 2024PodcastA 10-Person Team, One Click, and Two Years of Fake DeadlinesNotebookLM’s product and design leads explain how a source-grounded research tool became a viral two-host podcast generator—and what they refuse to add next.
59:30ENSequoia Capital· 8 oct. 2024PodcastGPT-4 Gets Half Your Data Questions Wrong. Snowflake's Answer Is Systems DesignUsing an off-the-shelf model on enterprise data yields about 45% reliability—one wrong answer out of every two.
45:22ENSequoia Capital· 2 oct. 2024PodcastAlphaGo Thought 30 Seconds a Move. o1 Made “Think Longer” GeneralOpenAI researchers Noam Brown, Ilge Akkaya, and Hunter Lightman on teaching language models to reason, the generator-verifier gap, and a second scaling curve at inference time.
39:45ENSequoia Capital· 24 sept. 2024PodcastHarmonic’s Bet: Teach AI Math Directly, and the Riemann Hypothesis Falls by 2029Robinhood’s Vlad Tenev and Harmonic co-founder Tudor Achim tell Sequoia why math is reasoning, why Lean is the verifier, and why they refuse to wait for mathematical skill to emerge from scale.
49:14ENSequoia Capital· 17 sept. 2024PodcastHumanoid robots fell from $1.5 million to $30,000. The GPT-3 moment is still a verbNvidia’s Jim Fan tells Sequoia how three kinds of data, a humanoid bet, and a System-1 motor-control model could unlock general-purpose robots.
51:15ENSequoia Capital· 10 sept. 2024PodcastAGI Has One Problem Left: Spend a Million Times More on a TokenMagic founder Eric Steinberger tells Sequoia’s Training Data why he trains his own models, why a software engineer is the MVP of AGI, and why 96% still is not a colleague.
1:12:31ENSequoia Capital· 27 août 2024PodcastA Bare Model Closes 35% of Airline Changes. Sierra's Agents Clear 70% of Real Tickets.Clay Bavor on why putting a raw LLM in front of customers does not math, and how AgentOS, supervisor agents, and outcome pricing change the arithmetic.
50:33ENSequoia Capital· 20 août 2024PodcastThe 40% data-center save started with an email that ended in a question markPhaidra founder Jim Gao tells Sequoia how an AlphaGo-inspired bet at Google became reinforcement learning that runs billion-dollar plants.
39:18ENSequoia Capital· 13 août 2024PodcastOpen and Closed Will Tie at the Same Size: PyTorch's Former Lead on Fast InferenceA six-month library swap at Meta became a five-year rebuild. Lin Qiao wants the rest of the industry to get that stack in five weeks.
1:07:34ENSequoia Capital· 6 août 2024PodcastGitHub Didn’t Believe Copilot Was Writing 25% of the CodeThomas Dohmke tells Sequoia how an unplanned GPT-3 experiment became Copilot, why GitHub still will not train its own foundation model, and how he wants a billion people writing software by 2030.
42:08ENSequoia Capital· 30 juil. 2024PodcastMeta 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.
51:36ENSequoia Capital· 23 juil. 2024PodcastKlarna's AI Took 700 Customer-Service Jobs. The Secret Was the ManualCEO Sebastian Siemiatkowski tells Sequoia how an OpenAI guinea-pig bet, a dispute-queue Slack message, and documentation written for an amnesiac employee produced $40 million of profit.
1:07:05ENSequoia Capital· 16 juil. 2024PodcastAlphaGo Combined Learning and Search. Language Models Only Did the First Half.Reflection AI's Misha Laskin tells Sequoia why prompted coding agents stall in the teens, and why the missing piece is still a ground-truth reward.
1:00:28ENSequoia Capital· 9 juil. 2024PodcastScaling Laws Aren't Done. Don't Lock Yourself Into Yesterday's Architecture.Microsoft CTO Kevin Scott on why the exponential is still compounding, why inference will dwarf training, and why a language model is the world's crappiest database.Charger plus24 sur 101 affichés