Latest Interviews
Showing 1–15 of 15 transcripts.
Clear all filters- Sequoia Capital22 min
Continual Learning: How AI Agents Get Better With Every Use | Arjun Karanam, Trajectory
Arjun Karanam, Ronak, Gabe, Harrison, Nico, Harvey
Trajectory, co-founded by Arjun and Ronak, addresses the lack of accumulated experience in AI by building a platform that enables models to continuously learn from the 100 trillion daily tokens generated by real-world agent interactions. The company utilizes a dual-learning architecture combining differential privacy with reinforcement learning on user-corrected traces, allowing organizations to transition from static models to systems that compound capability through automated post-training and flexible harness optimization. By abstracting complex training parameters into a 15-minute workflow, Trajectory empowers enterprises to retain ownership of their specialized models while refining agent performance directly against production traffic.
- Sequoia Capital17 min
How Companies Are Building Their Own Intelligence | Sonya Huang, Sequoia Capital
Approximately 80 portfolio company founders and AI leaders convened to strategize the adoption of Sovereign AI, a framework defined by vertical integration where organizations own model weights rather than relying on external APIs. The event combined high-level market analysis with technical workshops led by industry experts to outline a four-step roadmap for building custom intelligence capabilities. Participants explored critical architectural decisions regarding cost efficiency, latency reduction, and the necessity of dedicated research labs to leverage open-weight models for proprietary domain performance.
- Sequoia Capital9 min
Inside the Rise of Autonomous AI Hackers: XBOW's Oege de Moor
The presentation argues that the cybersecurity arms race has shifted to autonomous AI attacks, exemplified by an AI agent named XBO that recently achieved global dominance on HackerOne by discovering critical Microsoft Bing vulnerabilities through black-box testing. Because current defensive tools often fail to verify exploitability in live environments, the speaker urges organizations to immediately integrate autonomous AI into their workflows to counter negative exploit velocity before open-weight models close the capability gap within six to nine months. Ultimately, the event posits that future security success depends entirely on adopting AI-driven offensive and defensive systems rather than relying on traditional human-only methods.
- Sequoia Capital14 min
Why the Brain Computes 1,000,000x More Efficiently Than A GPU: Unconventional AI's Naveen Rao
Unconventional AI CEO Navin Rao is deploying a prototype that replaces traditional von Neumann architectures with nonlinear dynamical systems to overcome the impending energy saturation of current AI infrastructure. By leveraging Kuramoto synchronization models, the startup achieved functional generative capabilities in six months while demonstrating energy efficiency comparable to biological neural networks. This physics-based approach aims to bypass the thermodynamic limits of digital lithography, offering a viable pathway to artificial general intelligence within strict global power constraints.
- Sequoia Capital12 min
Starcloud's Philip Johnston: Why the Cheapest Compute Will Be in Space
StarCloud CEO Philip Johnston validated the technical feasibility of space-based high-performance computing through the successful "StarCloud 1" mission, which demonstrated thermal management and radiation tolerance while executing AI inference tasks. The company has filed an FCC application for an 88,000-satellite constellation capable of delivering 20 gigawatts of compute power with sub-50-millisecond latency, targeting a $100 billion capital expenditure that becomes economically viable once launch costs drop below $500 per kilogram. While current operations focus on inference workloads, the roadmap envisions future large-scale training structures that could catalyze a transition toward a Kardashev Type 2 civilization within decades.
- Sequoia Capital9 min
Why Data Is the Real AI Bottleneck: Flapping Airplanes' Ben and Asher Spector
Launched three months ago, Flapping Airplanes is an AI lab founded by Ben, Asher, and Thiel Fellow Aidan Smith that targets data-scarce domains like robotics and scientific discovery. The company differentiates itself through a system-level approach that builds custom hardware-abstraction layers to achieve theoretical 1,000x data efficiency, bypassing the limitations of standard frameworks like PyTorch. This strategy aims to democratize access to advanced AI by overcoming the escalating costs of data acquisition, prioritizing the recruitment of unconventional minds to drive paradigm shifts in system co-design.
- Sequoia Capital27 min
Waymo's Dmitri Dolgov: 20 Million Rides and the Road to Full Autonomy
Dmitri Dolgov, Konstantine Buhler
Dmitry Dolgov, Waymo's co-founder, outlines a 21-year evolution from DARPA challenges to a sixth-generation autonomous fleet that has achieved 200 million fully autonomous miles and delivered over 20 million rides with a safety profile 13 times superior to human drivers. The company's architecture utilizes a multimodal "Foundation Model" integrating physics reasoning and spatial understanding to power a structured, end-to-end system across 11 cities, now accelerating toward global expansion in London and Tokyo through the newly launched Waymo Origin vehicle. This mission-driven approach bypasses incremental improvements to rapidly scale commercial deployment, treating full autonomy as an existential imperative to eliminate the 13 million annual road fatalities while transitioning the core technology into a primary family transportation tool.
- Sequoia Capital20 min
Robotics' End Game: Nvidia's Jim Fan
NVIDIA Robotics is transitioning from language-centric VLA models to World Action Models like Dream Zero, which prioritize learning physical laws through video prediction to enable zero-shot generalization. This architectural shift is supported by a new data collection strategy replacing low-efficiency teleoperation with scalable egocentric video and neural simulators, establishing a log-linear scaling law between training hours and dexterity. Looking toward 2040, Jim Phan projects the industry will achieve a "Physical Turing Test," an orchestration "Physical API" for automated fleets, and "Physical Auto-Research" where robots autonomously improve their own designs.
- Sequoia Capital32 min
OpenAI’s Sam Altman on Building the ‘Core AI Subscription’ for Your Life
Sam Altman, Jens Nordvigsen, DAN GALPIN, SAM SACCONE
OpenAI has evolved from a 14-person research lab into a commercial powerhouse driven by the GPT-3 API and the viral ChatGPT platform, which now serves 500 million weekly active users. Under Sam Altman's leadership, the organization prioritizes high-impact small teams and a vision of becoming a personalized AI operating system while navigating a market where startups outpace legacy enterprises in agility. The company's roadmap emphasizes the transition from text-based assistants to autonomous agents by 2025, with future infrastructure expected to facilitate seamless agent-to-agent communication and eventual robotics applications.
- Sequoia Capital7 min
Open Evidence Captures Doctors’ Collective Wisdom with AI ft. Zachary Ziegler
OpenEvidence, a medical search tool used daily by over 25% of US physicians, recently enabled an internal medicine doctor to successfully manage an in-flight emergency for an immunocompromised cancer patient without diverting the aircraft. By synthesizing current literature and CDC data, the platform confirmed the patient did not require immediate intervention and provided a targeted treatment protocol that ensured a positive outcome upon landing. Building on this high-volume, real-time performance, the company is now expanding into direct workflow integration and developing an Aggregate Clinical Wisdom initiative to encode expert consensus from top specialists across various fields.
- Sequoia Capital10 min
9 Years to AGI? OpenAI’s Dan Roberts Reasons About Emulating Einstein
OpenAI demonstrated a strategic pivot toward test-time compute and reinforcement learning by showcasing models like O1 and O3 that solve complex physics problems, such as quantum electrodynamics and General Relativity, in minutes compared to human timescales. To support this new scaling paradigm where reasoning effort outweighs pre-training data, the organization plans to raise $500 billion to construct massive data center facilities in Abilene, Texas. This infrastructure investment aims to accelerate a timeline predicting that within nine years, systems will evolve from reproducing existing scientific knowledge to generating novel discoveries comparable to Einstein's major breakthroughs.
- Sequoia Capital18 min
The Physical Turing Test: Jim Fan on Nvidia's Roadmap for Embodied AI
NVIDIA Director Jim Phan outlines the "Physical Turing Test," a benchmark for robots performing complex physical tasks indistinguishable from human performance, while identifying data scarcity as the primary bottleneck for current systems. To overcome this, NVIDIA employs Simulation 2.0, utilizing generative models like Robocasa and video diffusion to exponentially scale training environments and introduce the GR00T N1 open-source model capable of zero-shot generalist control. This technological evolution aims to transition the physical world into a "Physical API," where software-defined skills enable autonomous robots to seamlessly handle household and industrial tasks, effectively embedding human-like dexterity into everyday infrastructure.
- Sequoia Capital29 min
AI's Trillion-Dollar Opportunity: Sequoia AI Ascent 2025 Keynote
Pat Grady, Sonia, Konstantin Vargas
Sequoia Capital outlines a strategic framework for the AI opportunity, predicting that the sector will disrupt both software and services markets by shifting value from selling tools to selling outcomes. The firm emphasizes investing in companies that demonstrate durable adoption and functional data flywheels, noting that 2024 marked a transition from hype to utility in vertical applications like healthcare and law. Looking ahead, the presentation predicts the emergence of an agent economy where interconnected agents manage resources and tasks, potentially enabling a "one-person unicorn" era through new management strategies.
- Sequoia Capital8 min
Ambient Agents and the New Agent Inbox ft. Harrison Chase
LangChain CEO Harrison Chase introduces ambient agents as background systems designed to execute complex, multi-step operations by monitoring event streams rather than relying on direct chat input. These agents operate within a strict human-in-the-loop framework that utilizes patterns like action approval, editing, and time-travel rollback to ensure accuracy while preventing full autonomy. To support this architecture, LangChain has enhanced its LangGraph infrastructure for state persistence and scalability while deploying an Agent Inbox UI that enables users to manage long-running workflows through direct oversight and feedback integration.
- Sequoia Capital14 min
What's next for AI agentic workflows ft. Andrew Ng of AI Fund
AI agents are driving a paradigm shift from single-step prompting to iterative workflows that combine reflection, multi-agent collaboration, tool use, and planning to achieve results that can surpass larger, faster models running in zero-shot mode. This approach allows systems using smaller language models like GPT-3.5 to outperform GPT-4 on complex benchmarks such as HumanEval by enabling self-correction loops and specialized role delegation. While reflection patterns are now robust enough for immediate integration, emerging capabilities in planning and multi-agent debate are expected to dramatically expand the scope of autonomous tasks over the coming year.