Derniers articles
4 471 articlesTrier
7:51ENRunning Guanaco 65B on a Cloud GPU with Runpod and Text Generation Web UIMatthew Berman · Forward FutureMatthew Berman demonstrates a 2023 workflow for renting a GPU, loading a quantized model and generating text. Model requirements, pricing and interfaces depend on the chosen version.
1:41:45ENThe Writers Aren't Luddites. They're Bargaining Over Who Holds the ModelsAnime : Nathan LabenzThree WGA members tell Nathan Labenz what ChatGPT can already do in a writers’ room — and which two studio nightmares the strike is trying to lock out.
1:07:30ENAdam Besvinick on Pre-Seed Investing and Raising a Small FundInvité : Adam Besvinick · Anime : Harry StebbingsThe Looking Glass Capital founder explains fund construction, LP relationships, thematic investing and a volatile fundraising market.
0:37ENTomer Cohen on How AI Changes Product WorkInvité : Tomer CohenA LinkedIn product leader compares the AI shift with the move from desktop to mobile.
8:26ENVoyager: How a GPT-4 Agent Learns Minecraft SkillsMatthew Berman · Forward FutureMatthew Berman explains a research system that proposes Minecraft tasks, stores reusable code and uses feedback to improve its next attempt.
2:07:04ENThe Ribosome Already Solved Digital ManufacturingInvité : Neil Gershenfeld · Anime : Lex FridmanNeil Gershenfeld tells Lex Fridman why Turing separated the head from the tape, how fab labs are the personal-computer moment for making things, and what it would take to grow a civilization from about twenty parts.
1:36:25ENTomer Cohen on LinkedIn Product Decisions and the AI ShiftInvité : Tomer Cohen · Anime : Harry StebbingsLinkedIn’s product chief discusses failed Stories, feed redesign, feedback practices and how AI could change the work of product teams.
0:59ENGenerative Agents as Characters in a Simulated WorldMatthew Berman · Forward FutureMatthew Berman introduces a Stanford and Google research architecture for language-model characters that interact in a sandbox.
1:20:49ENA Wafer-Scale Chip So AI Stops Waiting on AmdahlAnime : Nathan LabenzAndrew Feldman, CEO of Cerebras, tells Nathan Labenz why one 2.5-trillion-transistor die, sparse linear algebra, and a 16-machine supercomputer are a better shape than a GPU cluster.
7:37ENSmol Developer: Generating a Small Python Project From a PromptMatthew Berman · Forward FutureMatthew Berman walks through a 2023 project generator, from a simple Ruby example to a multi-file ChatGPT API program.
3:10:04ENTeachers keep getting fired for assigning Randall Kennedy's bookInvité : Randall Kennedy · Anime : Lex FridmanHarvard law professor Randall Kennedy tells Lex Fridman why he put English's most explosive slur on a book cover — and why blanking it, he argues, erases the history it carries.
1:18:41ENGuy Podjarny on Snyk, Product-Market Fit and Building Durable AdvantageInvité : Guy Podjarny · Anime : Harry StebbingsThe Snyk founder and angel investor explains how to reason about future markets, focus messaging and learn through iteration.
54:56ENKevin Scott Seized Microsoft's GPU Budget, Bet on OpenAI, and Still Didn't See ChatGPT ComingAnime : Sarah Guo, Elad GilFive months after ChatGPT launched, Microsoft's CTO walked No Priors through the all-in bet, the Copilot stack, and why he is still bullish on jobs.
5:49ENIan on Software Engineering at Jane StreetIan · Jane StreetA London desk developer explains code review, OCaml training, intern projects and the feedback that comes from sitting near users.
6:42ENEzio on Linux Engineering and Collaboration at Jane StreetEzio · Jane StreetA London engineer explains his team’s infrastructure work, the Pride community group and why new hires should ask questions.
4:51ENRachel on Trading Desk Operations at Jane StreetRachel · Jane StreetA London trading desk operations professional describes her daily work, collaboration, interview advice and experience of the firm's culture.
1:33:51ENGPT-4 still writes SQL that will not run. She put a guardrail on the modelAnime : Nathan LabenzShreya Rajpal built Guardrails AI so developers can encode what “correct” means, validate the pieces, and send failures back to the model instead of hoping the prompt holds.
1:29:33ENNico Wittenborn on Consumer Subscriptions and Solo InvestingInvité : Nico Wittenborn · Anime : Harry StebbingsThe Adjacent founder discusses retention, acquisition, fund construction and how evolving trends shape early-stage bets.
1:25:35ENTwo and a Half Years From 33.3% to 86.5%: Google Sent Foundation Models to Medical SchoolAnime : Nathan LabenzVivek Natarajan walks Nathan Labenz through Med-PaLM and Med-PaLM 2: the data-light techniques, the nine-axis clinician evals, and why a high USMLE-style score is the wrong trophy.
10:16ENDarkBERT: Adapting a Language Model to Dark-Web TextMatthew Berman · Forward FutureMatthew Berman reviews research on a RoBERTa-based model and its cybersecurity evaluations.
50:24ENRyan Wiggins on Growth Systems, WhatsApp and HiringInvité : Ryan Wiggins · Anime : Harry StebbingsMercury’s growth leader explains retention, experiments, small teams and the transition from product-market fit to repeatable growth.
1:38:22ENSix days after “slow down,” the plugins went liveAnime : Nathan LabenzConjecture CEO Connor Leahy talks GPT-4, the magic of black-box models, and cognitive emulation as a bounded alternative to god-like AGI.
42:18ENTen Percent of Search Is Health — and the Doctor Is Googling TooAnime : Sarah Guo, Elad GilGoogle researcher Karan Singhal on Med-PaLM 2, the real bar for medical language models, and why medicine is a harder alignment lab than chatbots.
59:38ENRichard Eisenberg on Dependent Types, OCaml, and AI-Assisted ProgrammingYaron Minsky, Richard Eisenberg · Jane StreetHow type systems can express correctness and performance, and why precise, readable specifications still matter when AI writes code.
Charger plus24 sur 4 471 affichés









