newsfilter.io

Latest Interviews

Showing 1–5 of 5 interview transcripts.

Clear all filters
  1. Sequoia Capital1h 0m

    OpenAI Sora 2 Team: How Generative Video Will Unlock Creativity and World Models

    Bill Peebles, Thomas Dimson, Rohan Sahai, Konstantine Buhler, Sonya Huang

    OpenAI's Sora leverages Diffusion Transformers and space-time tokens to generate coherent videos that increasingly respect physical laws, driving a daily output of seven million clips through an iterative deployment strategy. The platform prioritizes user creation over consumption via algorithmic incentives and the viral "Cameo" feature, while a dedicated API and emerging IP monetization frameworks support diverse enterprise and creator ecosystems. Looking forward, the technology aims to facilitate scientific discovery in physical sciences and evolve into a persistent digital clone platform for social interaction and knowledge work.

  2. Sequoia Capital55 min

    Using AI to Empower Creators fr Roblox Studio Head Stef Corazza

    Stef Corazza, Konstantine Buhler, Sonya Huang

    Roblox is leveraging its $29 billion market-cap infrastructure and a multimodal dataset authorized by its creator community to deploy generative AI tools that increase developer output by up to 180%. These features, including code assistants and 3D material generators, are currently driving a 30% higher game publication rate while maintaining safety through strict moderation protocols. The company's roadmap aims to transition game creation from pixel-level control to intent-based design, utilizing 3D foundational models to enable dynamic storytelling and real-time visual personalization.

  3. Sequoia Capital52 min

    Turning Graph AI into ROI ft Kumo’s Hema Raghavan

    Hema Raghavan, Konstantine Buhler, Sonya Huang, Constantine, Sonia

    Kumo AI co-founder Hema Raghavan presents a GPU-accelerated AutoML platform that automatically constructs graph structures from relational data to enable predictive SQL-like queries without manual engineering. The system delivers rapid four-week proofs of concept for sectors ranging from fintech fraud detection to healthcare demand forecasting by integrating directly with Snowflake and Databricks while maintaining strict data residency. By combining optimized cost architectures with explainability features and LLM synergies, Kumo lowers the barrier to graph learning for diverse enterprises requiring immediate, accurate model insights.

  4. Sequoia Capital52 min

    Cracking the Code on Offensive Security With AI ft XBOW CEO and GitHub Copilot Creator Oege de Moor

    Oege de Moor, Konstantine Buhler, Sonya Huang

    Former Oxford professor and GitHub Copilot creator Uge Demore's company Expo deploys an autonomous AI system to automate offensive security testing, achieving an 85% success rate on proprietary benchmarks while matching top human penetration testers in 28 minutes instead of 40 hours. The platform continuously identifies critical vulnerabilities in major financial institutions and replaces traditional $18,000 manual tests with a scalable, subscription-based service designed to outpace AI-assisted cyber threats. Operating on a foundation of proprietary training data and strict cloud-based guardrails, Expo aims to transform web security standards by making continuous, automated offensive testing accessible to organizations of all sizes.

  5. Sequoia Capital1h 3m

    Getting the Most From AI With Multiple Custom Agents ft Dust’s Gabriel Hubert and Stanislas Polu

    Gabriel Hubert, Stanislas Polu, Konstantine Buhler, Pat Grady

    Dust positions itself as a horizontal platform for AI adoption, predicting a bimodal future where enterprise users seamlessly switch between frontier APIs and local models to navigate varying technological plateaus. By prioritizing product-market fit over proprietary model training and leveraging Retrieval-Augmented Generation to unlock data silos, the company enables diverse teams to build specialized agents that augment human work rather than replace it. This strategy targets a demographic of young power users and aims to scale from isolated pilots to organization-wide adoption, facilitating everything from cross-functional translation to global expansion despite current limitations in reasoning breakthroughs.