Latest Interviews
Showing 61–75 of 164 transcripts.
Clear all filters- RAISE Summit18 min
CJ Desai, MongoDB & Laura Modiano, OpenAI in Conversation | RAISE Summit 2026
MongoDB, a cloud-native database serving 67,500 customers including JPMorgan Chase, differentiates itself as the only hybrid multi-cloud solution capable of unifying vector data, operational records, and auto-embeddings to support AI-native architectures. As enterprises transition toward 2026 production-grade agentic systems, the platform addresses critical scaling bottlenecks by offering schemaless flexibility that accommodates compressed AI model iteration cycles of just six weeks. This strategic positioning enables organizations to move beyond prototyping by solving data sovereignty concerns while accelerating innovation through forward-deployed engineering support and a unified data layer that eliminates the need for separate vector stores.
- RAISE Summit7 min
Who Controls AI? The Open Source Imperative | Arthur Mensch, Mistral AI | RAISE Summit 2026
Arthur Mensch, Mark Surman, Henri Delahaye
Speakers Arthur and Marc outline a strategic framework for artificial intelligence sovereignty that prioritizes domestic compute security, affordability, and sustainability to mitigate economic inequality and social unrest. They identify open source technology as a critical mechanism for nations to retain control over essential infrastructure, preventing foreign dependency while fostering a global ecosystem where open components serve as the dominant building blocks within the next decade. Although this shift from proprietary chatbots to shared infrastructure is widely anticipated, the speakers warn that realizing this future requires sustained research funding and coordinated interoperability to overcome significant capital and partnership challenges.
- RAISE Summit22 min
AI Is the New Operating Environment. Data Is the New Control Plane. | RAISE Summit 2026
The speaker identifies that the AI arms race hinges on compressing time across token generation, environment deployment, and data orchestration. By replacing legacy 2D client-server protocols with a Linux kernel-native 3D parallel file system, the technology virtualizes existing storage assets to slash deployment timelines by a factor of 70 and eliminates bottlenecks during GPU training. This architectural shift not only mitigates current SSD shortages through efficient hybrid HDD-NVMe and tape strategies but also directly reduces capital and operational expenditures in high-performance computing environments.
- RAISE Summit14 min
Rewiring the Web: Data Access, AI Agents & the Next Digital Revolution | RAISE Summit 2026
Juras Juršėnas, Fred de Villamil, Chet Haase, Romain Guyett, Francois Chollette, Francisco Neri, Gérald
The event analyzes how the AI era has intensified the demand for live web data while rendering traditional scraping techniques obsolete due to sophisticated anti-bot defenses and shifting regulatory landscapes. It highlights critical economic tensions between content publishers and AI developers, arguing that the current ad-revenue model is threatened by AI aggregators that bypass source websites without compensation. Strategic recommendations focus on transitioning web infrastructure to be optimized for AI agents and establishing new frameworks for data monetization to ensure sustainable value distribution in the emerging agentic commerce ecosystem.
- RAISE Summit45 min
Startup Competition Final: RAISE the STAKES | RAISE Summit 2026
Gleb Budman, Cyrille Saint Olive, Luke Pappas, Shahab Shafieehanjani, Cyril, Rodrigo, Kinga Matua, Charles
Raise Summit's Raise the Stakes 2026 competition attracted 750 founders from 79 nations who collectively hold $800 million in valuation, with the event judged by executives from Backblaze, Google for Startups, and NEA. The jury selected CurryGen as the top winner for its proprietary biological data workflows, while Neural Trust and Strawberry Browser secured second and third places respectively for their enterprise security and agentic browser platforms. This cohort of finalists represents a high-growth segment of the AI ecosystem that has already raised $56 million and is poised to address urgent market needs in security, drug discovery, and workflow automation.
- RAISE Summit19 min
Fireside Chat with Parag Agrawal, Founder & CEO of Parallel Web Systems | RAISE Summit 2026
Parag Agrawal, Karim Jalbout, Vineet Bhanwajani, Vineet Bhanuri, Vineeth Pandey, Vikas Khannabiswamy
Parallels founder Vikas Khannabhiswamy redefined the scale of AI web usage, forecasting agents will consume data 100,000 to 1,000,000 times more than humans by deploying a specialized infrastructure layer for high-volume knowledge work. The platform addresses critical limitations of standard LLMs by integrating real-time web search and applying Shapley values to automate fair compensation for content owners, thereby disrupting traditional web monetization models. This approach targets sectors like AI law, science, and coding while establishing a new revenue framework where publishers are paid proportionally to their specific data contribution.
- RAISE Summit20 min
AI Compute: Rebuilt for the Agentic Era | Mark Papermaster, AMD | RAISE Summit 2026
Mark Papermaster, Karim Jalbout
Driven by a surge in AI adoption that propelled its market capitalization from $200 billion to nearly $1 trillion, AMD is targeting server CPU market leadership through a holistic system design philosophy that includes the upcoming 72-node "Helios" rack and two-nanometer "Venice" processors. The company has transitioned from component manufacturing to integrated rack-level solutions by acquiring ZT Systems and deploying open-standard AI stacks across major European supercomputers to address data sovereignty and energy efficiency demands. Underpinning this growth is a scrappy corporate culture focused on customer collaboration, which has enabled a strategic shift toward agentic AI workflows and expanded training for its 33,000-person workforce to align with enterprise needs.
- RAISE Summit25 min
AI, Security & Economic Power: A New Playbook for Democracies | Sebastian Kurz | RAISE Summit 2026
Sebastian Kurz, Jacob Helberg, Sebastian Matthes
At the second Pax Silicon Summit in Washington, European nations acceded to the Paxilica initiative while leaders like Jacob and Sebastian debated the viability of sovereign AI against a backdrop of historic U.S. economic growth driven by an AI boom. The discussion highlighted a strategic divergence where the U.S. promotes an ecosystem-based approach to sovereignty and deregulation, contrasting with European concerns over the Digital Markets Act and the high costs of total vertical integration. Ultimately, experts emphasized that while foundational model dominance remains difficult for Europe, securing critical infrastructure and shifting toward competitiveness are essential to prevent total reliance on superpowers or losing the financial capacity for social welfare.
- RAISE Summit20 min
Qasar Younis, Applied Intuition & Mamoon Hamid, Kleiner Perkins | RAISE Summit 2026
Founded by automotive veterans Mamoun Lachheb and Peter, Applied Intuition pivoted from a robotic taxi ambition to a self-funding platform generating revenue from commercial L4 trucking in Japan and a multi-hundred million dollar defense sector. The company now distinguishes "Physical AI" through end-to-end transformer models that enable dynamic human-machine collaboration across diverse vehicle form factors while maintaining radical pragmatism by keeping 93% of its $1 billion in capital on the balance sheet. This strategy supports a culture of radical pragmatism where over 1,000 engineers prioritize profitability and survival, fostering a "founders mafia" of 40 alumni who spin out new ventures using the company's shared infrastructure.
- RAISE Summit20 min
Fireside Chat with Ekin Dogus Cubuk, Co-CEO of Periodic Labs | RAISE Summit 2026
Founded by former Google Brain and DeepMind researchers, a new company is deploying a hybrid infrastructure of large language models, reinforcement learning, and high-throughput robotics to overcome data limitations in physical science. The Bay Area laboratory automates semiconductor and materials research by executing thousands of daily experiments to create a recursive improvement cycle for chip manufacturing and solid-state chemistry. With an immediate goal of reaching 1,000 experimental trials per day, the venture aims to generate rapid process engineering breakthroughs for the industry while evolving toward profitable intellectual property generation.
- RAISE Summit21 min
From AI Adoption to AI ROI | Glean & Altimeter Capital | RAISE Summit 2026
Arvind Jain, Apoorv Agrawal, Vikas Khannabiswamy, Shyam Gollakota, Vikram Bajaj
Glean, co-founded by former Google and Rubrik executive Arvind Krishnan, differentiates itself in the enterprise AI market by building a proprietary context graph and retrieval layer rather than developing foundation models. The company is strategically pivoting from personal productivity tools to measurable departmental use cases like customer service and legal review to demonstrate clear return on investment amid rising infrastructure costs. By maintaining neutrality across major model providers and utilizing limited forward-deployed engineering, Glean aims to solve the critical gap between raw AI capabilities and secure, accurate enterprise implementation.
- RAISE Summit21 min
The New AI Challenge: Trust, Control, and Scale | JetBrains x NVIDIA | RAISE Summit 2026
As enterprises transition from individual AI literacy to team-level agentic deployment, a strategic shift is required to restructure engineering units into integrated "teams of builders" that maintain shared context across the product lifecycle. This evolution necessitates a three-layer governance framework combining individual tool integration, centralized platform policy enforcement, and collaborative team environments to ensure agentic intelligence remains observable, traceable, and auditable. By leveraging open standards like the Agent Communication Protocol and secure infrastructure solutions, organizations can mitigate risks such as shadow AI and prompt injections while focusing on extracting tangible business value from agentic inference rather than superficial metrics.
- RAISE Summit21 min
Crusoe's Energy-First Vision for the AI Cloud | Chase Lochmiller | RAISE Summit 2026
Chase Lochmiller, Henri Delahaye, Francesc Campoy, DAN GALPIN, FRANCOIS BEAUFORT
Crusoe is establishing itself as a vertically integrated AI infrastructure leader by relocating compute from traditional hubs to stranded energy sources, having secured over 40 GW in its development pipeline with rapid construction cycles like the 11-month Abilene project. The company recently acquired Atero to deploy its "Crusoe Memory Alloy" for optimized inference, simultaneously shifting its product strategy toward distributed "AI factories" that monetize hardware investments through cost-effective token generation rather than large-scale training. As the industry transitions from theoretical announcements to execution over the next 24 months, Crusoe prioritizes operational resilience and end-user value delivery to dominate the emerging market for decentralized, energy-driven AI services.
- RAISE Summit19 min
Building AI That Gets Work Done | Gautier Cloix, H Company | RAISE Summit 2026
CEO Gautier has transformed the enterprise AI landscape by launching H, a solution that bypasses legacy system barriers through third-generation "computer use" technology capable of mimicking human actions on any software interface. This approach delivers measurable financial returns in weeks rather than months, evidenced by 4–5% procurement savings in construction and streamlined operations for major clients like BNP Paribas. By leveraging cost-effective, proprietary models within Nvidia's ecosystem, H competes globally by prioritizing immediate balance sheet impact over theoretical capabilities, effectively removing the execution layer to let human capital focus on strategic intent.
- RAISE Summit22 min
Fireside Chat with Lin Qiao, CEO & Co-Founder of Fireworks AI | RAISE Summit 2026
At the RAISE conference, Matt Miller of Avantik VC discussed Fireworks AI's specialized intelligence platform with founder Lin Chao, a former Meta AI infrastructure lead who built the company to help businesses own their data-driven models rather than relying on generic APIs. The dialogue highlighted how fine-tuning open-source models on proprietary data allows companies to reduce inference costs by up to tenfold while creating durable competitive moats against commoditized application development. Together, they outlined a strategic shift where long-term business survival depends on continuous model customization and the deployment of millions of specialized models tailored to specific use cases.