Sequoia Capital

•

Kanál • Pořady: 3

Technologie

Technologie

Týdně
Firmy

Firmy

Každé dva týdny
Podnikání

Podnikání

Každé dva týdny

před 6 dny

Why Hardware-Software Co-Design Is AI's Real 100x: Dylan Patel of SemiAnalysis

Dylan Patel, founder of SemiAnalysis, argues the biggest gains in AI don't come from faster chips, they come from software-hardware co-design. Optimizing the model, the kernels, and the silicon together turns a 2x here and a 2x there into 100x. He explains why DeepSeek's experts were shaped for Nvidia's Hopper (and why TPUs struggle to run it), why OpenAI's sparser models and Anthropic's denser ones pull them toward different hardware, and why the so-called CUDA moat was never really about CUDA. Dylan breaks down InferenceX, his living benchmark that runs the latest models on over $50M of donated hardware daily, tracking a roughly 60x annual drop in cost per unit of quality. He makes the case that inference will be a bigger market than oil, that the compute crunch persists because models expand the value of useful work faster than compute grows, and why Jensen Huang is bankrolling neoclouds to engineer a multipolar world. Hosted by Shaun Maguire and Sonya Huang, Sequoia Capital

před 6 dny

•

1 h 10 min
30. 4.

Andrej Karpathy: From Vibe Coding to Agentic Engineering

Andrej Karpathy (co-founder of OpenAI, former head of AI at Tesla, and now founder of Eureka Labs) talks with Sequoia partner Stephanie Zhan at AI Ascent 2026 about what's changed in the year since he coined "vibe coding." He explains why he's never felt more behind as a programmer, why agentic engineering is the more serious discipline taking shape on top of vibe coding, and why we should think of LLMs not as animals but as ghosts: jagged, statistical, summoned entities that require a new kind of taste and judgment to direct. He also touches on Software 3.0, the limits of verifiability, and why you can outsource your thinking but never your understanding.

30. 4.

•

30 min
19. 5.

Rebuilding IT From the Ground Up for the AI Age: Serval's Jake Stauch

Jake Stauch, founder and CEO of Serval, is building a ServiceNow for the AI era. His most contrarian bet is that the product should look like boring old enterprise software, but with unlimited intelligence. Serval's architecture splits work between two agents: an admin agent that uses code generation to spin up workflows from natural language, and a help desk agent that can only act through the tools admins explicitly approve. Jake explains why his team uses OpenAI models for end-user interaction and Anthropic models for code generation, why new model releases sometimes have to be rolled back when prompt tuning breaks, and why he's not worried the foundation labs will come downmarket. He also makes the case for "fewer, better" hiring as the only durable moat in a world where products may need to be rebuilt every six months. Hosted by Pat Grady, Sequoia Capital

19. 5.

•

38 min
2. 4.

Jack Dorsey: Every Company Can Now Be a Mini-AGI

Jack Dorsey (Block CEO) and Roelof Botha (Sequoia partner and Block board member) join to discuss a bold claim they wrote about recently: the traditional corporate hierarchy isn't just inefficient — it's obsolete. Jack made one of the toughest calls in recent business history: cutting 40% of his workforce and rebuilding the company from the ground up around what he calls an AI "intelligence layer." We get into how that conversation went down, the math they used to land on a number, and why he's convinced that acting from a position of strength beats reacting from one of weakness. Jack breaks down his vision for simplifying into just three roles, and what it means to replace a pyramid org chart with a circle — AI at the center, and people at the edge. Roelof, who helped think through the restructuring, shares his perspective on how AI-native startups are building differently, and what CEO qualities are timeless. I've eliminated org charts before. I know how hard this is. But Jack is doing something I never had the tools to pull off. If you're a founder wondering whether your hierarchy is working for you or against you, this one will make you uncomfortable in the best way.

2. 4.

•

1 h 4 min
5. 5.

Anthropic's Boris Cherny: Coding's Printing Press Moment

Boris Cherny, creator of Claude Code at Anthropic, joins Sequoia partner Lauren Reeder at AI Ascent 2026 to talk about where coding goes from here. He explains why he hasn't written a line of code in 2026, why he now ships dozens of PRs a day from his phone, and why he believes coding is effectively solved — at least for the code he writes. Also: why loops are the future, why he thinks Claude Code itself may be 100 lines of code a year from now, and why the invention of the printing press is the right analogy for what's about to happen to software.

5. 5.

•

25 min
26. 5.

How Cursor Trained Composer on Fireworks: Distributed Infrastructure for High-Performance RL

Cursor's Federico Cassano and Fireworks' Dmytro Dzhulgakov explain how they collaborated to build Composer as a specialized foundation model. The core insight: models have finite capacity in their weights, and allocating all those bits to the singular task of software engineering in Cursor frees the model to be both better at the task and far more efficient at inference. Rather than start from pre-training and work up, they took an unconventional top-down approach — mid-training and RL on top of an open-source base to get a useful model into users' hands fast, then specializing the model around real Cursor usage. With Fireworks providing distributed infrastructure, Composer delivers frontier-class coding performance with the speed of a much smaller model. Hosted by Sonya Huang, Sequoia Capital

26. 5.

•

46 min
24. 3.

How Autonomous Labs Will Transform Scientific Research: Ginkgo Bioworks’ Jason Kelly

Jason Kelly founded Ginkgo Bioworks in 2008 with a simple but radical idea: DNA is code, and cells are programmable. Sixteen years later, AI is finally making that vision real in ways that could reshape science itself. Jason describes a landmark collaboration with OpenAI in which a reasoning model with access to a robotic lab beat the state of the art in biochemistry by 40% - not by being smarter than scientists, but by running experiments 24 hours a day and sharing data across a hundred parallel hypotheses simultaneously. He argues that the biggest inefficiency in science isn't intelligence, it's manual labor. Once AI helps scale research, the cost of discovery collapses and breakthroughs follow, with profound implications for biopharma, national competitiveness, and human health. Hosted by Sonya Huang and Pat Grady, Sequoia Capital

24. 3.

•

58 min
21. 1.

Context Engineering Our Way to Long-Horizon Agents: LangChain’s Harrison Chase

Harrison Chase, cofounder of LangChain and pioneer of AI agent frameworks, discusses the emergence of long-horizon agents that can work autonomously for extended periods. Harrison breaks down the evolution from early scaffolding approaches to today's harness-based architectures, explaining why context engineering - not just better models - has become fundamental to agent development. He shares insights on why coding agents are leading the way, the role of file systems in agent workflows, and how building agents differs from traditional software development - from the importance of traces as the new source of truth to memory systems that enable agents to improve themselves over time. Hosted by Sonya Huang and Pat Grady

21. 1.

•

40 min
23. 09. 2025

Why Businesses Are Rejecting the AI They’ve Asked For: Agency CEO Elias Torres

Elias Torres has been building AI systems since 1999, from chatbots at IBM to co-founding Drift and now Agency. He believes businesses are caught in an expectation mismatch—demanding AI while rejecting it due to imperfection anxiety. Drawing from his experience scaling HubSpot, Elias explains why human-led customer experience doesn’t scale and how Agency is building AI-first solutions that work autonomously. His contrarian approach focuses on the back-end customer experience rather than front-end AI SDRs, aiming to “deprogram the entire business world” from inefficient human-dependent processes. Hosted by: Sonya Huang and Pat Grady, Sequoia Capital Mentioned in the episode: Lookery: David Cancel’s first startup that Elias joined after IBM; shut down in 2009 Performable: Elias and David’s second startup, acquired by HubSpot in 2011 Drift: Elias and David Cancel’s third startup, merged with Salesloft in 2024 Klaviyo: B2C CRM company started by Andrew Bialecki after working with Elias at HubSpot Secret: Short-lived anonymous messaging app that inspired one of Drift’s early iterations Tatajuba: Kitesurfing destination in Jericoacoara, Brazil where Elias (briefly) considered retirement 00:00 Introduction 01:50 AI and Customer Expectations 03:36 Managing Emails with AI 07:21 Elias' Personal Journey 11:27 Early Career 14:28 Joining HubSpot and Scaling Challenges 16:31 Hiring Exceptional Talent 18:53 Founding Drift 20:27 Pivoting to Success with Drift 21:41 Drift's Chatbot Innovation 22:09 Challenges and Limitations of Drift 22:37 The Struggle with Customer Knowledge 23:09 Scaling Challenges and Lessons Learned 25:58 Rediscovering Purpose Post-Drift 28:55 The Birth of Agency 29:42 AI's Role in Customer Experience 35:13 Building a Sustainable Business Model 37:06 The Vision for Agency 38:22 Challenges and Opportunities with AI 41:22 Deprogramming and Embracing Change 43:23 Optimism for the AI Future 44:15 Closing Thoughts

23. 09. 2025

•

45 min
14. 4.

From SEO to Agent-Led Growth: Profound’s James Cadwallader

James Cadwallader, co-founder and CEO of Profound, makes the case that we are living through the biggest platform shift in marketing history. The front door of the internet hasn't changed, but the visitor walking through it has. Where consumers once clicked blue links, AI agents now crawl the web on their behalf, synthesizing answers and steering purchase decisions at scale. James explains why Gemini, ChatGPT, and Claude all recommend brands differently, why mapping AI visibility onto traditional SEO is the wrong instinct, and why the real imperative is to equip a superintelligent agent with original insight it couldn't find anywhere else. He also digs into the dead internet theory – the possibility that human browsing could largely cease within three years – how AI advertising may become the most effective form the world has ever seen, and why agent-led marketing isn't just automation of the old work, but an entirely new capability. Hosted by Sonya Huang, Sequoia Capital

14. 4.

•

31 min

We help the daring build legendary companies – from idea to IPO and beyond.

Webová stránka