Results
17 articlesSort
1:50:48EN
Y Combinator· Oct 24, 2026Why Inference Matters 12× More Than TrainingFrançois Chaubard, Dan Fu, Jeff Morgan, Joe Fioti, Liangsheng Yin · Y CombinatorYC Paper Club walks the inference stack from an API request down to GPU kernels. Four talks from Together AI, Ollama, Luminal, and SGLang argue that serving quality, not just training FLOPs, is where most of the compute and the money now sit.
38:14EN
No Priors: AI, Machine Learning, Tech, & Startups· Sep 18, 2026Podcast · 2 speakersDiffusion Will Win Inference: Stefano Ermon Bets Parallel Tokens Beat Custom SiliconHost Sarah Guo, Elad GilThe Stanford diffusion pioneer tells No Priors why a 50-person lab is training language models that generate many tokens at once — and why that maps to GPUs the way transformers once mapped to training.
1:10:15EN
Sequoia Capital· Jun 30, 2026PodcastThe 100x lives between the layers: Dylan Patel on co-design and a 60x yearSemiAnalysis’s founder argues that matching models, kernels, and silicon is what turns a 2x here and a 2x there into 100x, while equivalent-quality inference has gotten about 60 times cheaper in a year.
59:06EN
Unsupervised Learning: With Jacob Effron· Sep 10, 2025PodcastInference Costs Fell 100x. Tri Dao Says There's Another 10x — and NVIDIA's 90% Won't HoldHost Jacob EffronThe FlashAttention author and Together AI chief scientist on why hardware portability is a myth, how Claude Code made him 1.5x faster, and why workloads will split the chip market.
46:35ENSequoia Capital· Nov 13, 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.
36:49ENNo Priors: AI, Machine Learning, Tech, & Startups· Nov 7, 2024Podcast · 2 speakersJensen Huang: NVIDIA Is No Longer Building Computers. It Is Building AI FactoriesHost Sarah Guo, Elad GilOn No Priors, NVIDIA’s CEO walks through hyper Moore’s law, NVLink, a 100,000-GPU cluster stood up in weeks, and why tokens are the new commodity.
39:18ENSequoia Capital· Aug 13, 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.
1:52:18EN
Cognitive Revolution "How AI Changes Everything"· May 6, 2024PodcastTwo Weeks, Thirty Compiles: How Adam Majmudar Wrote a GPU From ScratchHost Nathan LabenzA Cognitive Revolution walkthrough of TinyGPU, from SIMD kernels and memory latency to an open-source chip flow that runs on a laptop.
1:42:17ENCognitive Revolution "How AI Changes Everything"· Apr 20, 2024PodcastThe Cheapest Inference Provider Is the Most DangerousHost Nathan LabenzFireworks AI cofounders Lin Qiao and Dmytro Ivchenko explain why subsidized GPU bills vanish, how LoRA serving can cut cost a hundredfold, and why OpenAI is the rival that actually matters.
38:33ENNo Priors: AI, Machine Learning, Tech, & Startups· Mar 21, 2024Podcast · 2 speakersSpeed Is the Thing You Can Buy: Baseten’s Bet on InferenceHost Sarah Guo, Elad GilTuhin Srivastava tells Sarah Guo and Elad Gil why a six-person chatbot company running a billion tokens a day should not be building its own serving stack.
2:30:06ENCognitive Revolution "How AI Changes Everything"· Dec 22, 2023PodcastTrained on a Thirtieth of GPT-4's Tokens, Mamba Beat Transformers on Home TurfHost Nathan LabenzIn a solo emergency episode, Nathan Labenz argues that Albert Gu and Tri Dao's selective state space model is the first architecture since 2017 that deserves to share the Transformer’s throne.
3:08:46EN
Lex Fridman· Jun 29, 2023Podcast · 2 speakersWho owns the off switch? George Hotz on tinygrad, open-source AI, and GodGuest George Hotz · Host Lex FridmanOn his third Lex Fridman podcast, the comma.ai and tiny corp founder argues that compute, intelligence, and the off switch should live in people's houses, not in a trusted few.
3:34:03ENLex Fridman· Jun 2, 2023Podcast · 2 speakersPython Isn't Finished: Chris Lattner on Mojo and the Hardware MessGuest Chris Lattner · Host Lex FridmanThe LLVM creator returns to Lex Fridman to argue that Python should grow C-like performance instead of forcing people out of the language.
52:11ENNo Priors: AI, Machine Learning, Tech, & Startups· Apr 20, 2023Podcast · 2 speakersJensen Huang: For 25 Years, 99 Percent Were Right. NVIDIA Bet on the 1 PercentHost Sarah Guo, Elad GilThe NVIDIA CEO tells No Priors how CUDA survived a decade without customers, why the company tries to stay lazy, and what he wants built for drugs and climate.
1:44:11ENLex Fridman· Aug 27, 2019Podcast · 2 speakersA Single GPU Is Enough: Jeremy Howard on Teaching Deep LearningGuest Jeremy Howard · Host Lex FridmanThe fast.ai founder tells Lex Fridman why domain experts, not more papers or more TPUs, will decide whether deep learning actually helps people.
1:01:23EN
Citadel SecuritiesPodcastJensen Huang on AI Factories, Digital Workers and Why the Build-Out Has Barely BegunNvidia's CEO tells a room of institutional investors how accelerated computing led to AI factories, and where agentic AI, robotics and sovereign AI go next.
17:19ENLex FridmanPodcast · 2 speakersFrom Atoms to Data Centers: How a Modern Processor Finds Parallelism in a Serial ProgramGuest Jim Keller · Host Lex FridmanJim Keller walks through the layers of abstraction in computer design, from transistors to out-of-order execution, branch prediction and deterministic answers.