Latest Interviews

Lambda, Hugging Face, SemiAnalysis: The Neural Network Is the Computer

The discussion outlines a fundamental shift from traditional programming to "neural software," where generative models act as dynamic interfaces that execute user intent through natural language rather than static code. Key developments in video, audio, and multimodal integration are enabling consistent, real-time content creation while replacing fixed graphical user interfaces with adaptive, semantic environments. This evolution promises to eliminate conventional coding barriers by 2030, though it simultaneously introduces complex challenges regarding state retention and the societal implications of AI-driven interactions.

Jaspar Carmichael-Jack, Co-Founder & CEO, Artisan: How We Generate 1000s of Leads with AI

Jaspar Carmichael-Jack, Jasper

Founded by CEO Jasper in September 2023, Artisan has rapidly scaled to over $6 million in annual recurring revenue as the second-fastest-growing AI startup globally following its Y Combinator Winter '24 graduation. The company distinguishes itself through a controversial "Stop Hiring Humans" marketing campaign that generated over one billion impressions and a proprietary AI agent named Ava, which recently delivered nearly $1 million in ARR from a $20,000 investment by autonomously managing outbound sales cycles. Looking ahead, Artisan is preparing a major Q4 product expansion while refining its strategy to leverage AI-generated content for high-volume marketing, reserving human oversight for SEO and brand-critical communications.

Alex Ratner, CEO & Co-Founder, Snorkel AI: Data Centric AI in the Agentic Era

Alex Ratner

Alex, co-founder of Snorkel.ai, argues that specialized agentic systems require robust evaluation benchmarks composed of prompt datasets, rubrics, and verifiers to serve as functional product requirements. His company provides data operations platforms to major US banks and federal agencies while demonstrating that custom-trained models using programmatic evaluators can outperform GPT-4 on complex insurance underwriting tasks. This approach establishes high-quality data and verification logic as the foundational prerequisites for optimizing Reinforcement Learning in enterprise environments.

Dr. Ilya Burkov (Nebius): Powering Next‑Gen Biotech Research with High‑Performance AI and Cloud GPUs

Dr. Ilya Burkov

Nebius, a vertically integrated Dutch cloud provider, leverages its unique pay-as-you-go infrastructure and top-tier supercomputer ranking to support biotech and healthcare firms in overcoming hardware bottlenecks. The platform enables critical advancements in drug discovery and radiology diagnostics for organizations like Converge Bio and Simulacra AI, which achieved significant accuracy gains and speedups over hyperscaler alternatives by utilizing on-demand H100 GPU capacity. Looking toward 2025, the company aims to expand its global footprint by integrating agentic AI and multi-omics data processing to address workforce shortages and accelerate time-to-market for new treatments.

Shaun O’Meara, CTO @ Mirantis: All Infrastructure is AI Infrastructure

Shaun O’Meara

Addressing critical constraints such as GPU scarcity and the need for unified management of hybrid AI and legacy workloads, Mirantis launched Cordant as an open-source platform designed to deliver composable, multi-cloud infrastructure. The solution utilizes a declarative, pattern-based approach to enable performance contracting and seamless resource orchestration across edge, on-prem, and bare-metal environments. By overlaying open-source management on strategic vendor hardware, the platform aims to eliminate single-vendor lock-in while providing an "AI Factory" reference architecture for accelerated deployment.

OVH Cloud, ClickHouse, Neon, Aiven, Airbyte & Commit: Open Software Infrastructure 2.0

Abel, German, Oskari, Heike, Michel, Alexei, Francois Beaufort

Executives from Aiven, Neon, Airbyte, ClickHouse, and OVHcloud converged to discuss how open-source data platforms are securing $10 billion in combined valuations while addressing enterprise needs for data sovereignty and interoperability. The panel detailed how the AI boom is driving a shift from human-led development to agentic provisioning, creating demand for storage-centric architectures that support short-lived, unstructured data workloads. Looking ahead, industry leaders predict a five-year trajectory where hundreds of millions of citizen developers leverage commoditized AI to compress development cycles, necessitating hybrid infrastructure strategies that balance local model execution with scalable cloud data management.

Yuri Frayman, Co-Founder & CEO, Cast AI: Robots Will Feed the World… If We Build the Rails

Yuri Frayman

SoftBank Vision Fund portfolio companies are constructing the physical infrastructure necessary to transition AI from knowledge work into the global economy, aiming to replicate the prosperity seen with previous industrial revolutions. While addressing critical constraints in energy and compute, these initiatives prioritize workforce re-education and decentralized, open-source architectures to navigate regulatory fragmentation and ensure fair access against well-funded commercial labs. The deployment of large-scale, cloud-agnostic inference networks over the next decade will ultimately determine which entities successfully harness this AI abundance.

Alix, SAP, Daytona, Feedly, Prosus & Handelsblatt: The Future Agentic AI & Autonomous Workflows

Alix, Edwin Kodabachan, Paul Van der Boor, Alexander Maisor, Emmanuel Casimates, Ivan Bourazin

Experts at a recent industry gathering defined Agentic AI as a shift from simple request-response models to autonomous systems capable of multi-step planning, emphasizing that security and governance remain critical barriers requiring strict human oversight. SAP and other leaders proposed frameworks like "agentic quality controls" and locally hosted environments to manage agent identities and comply with regulations, while anticipating a workforce evolution where humans supervise "digital teammates" rather than performing manual tasks. Despite concerns over the hype cycle, participants agreed that organizations with mature data and processes will soon leverage these tools to exponentially increase productivity, provided they implement robust guardrails to direct this technological capability toward tangible business value.

Karl Havard, COO, Nscale on 3 Things Sovereign AI Must Deliver: Control, Compliance, and Growth

Karl Havard, Carl

Carl, CCO of Nscale, defines "sovereign AI" as a full-stack alternative to US hyperscalers that integrates energy generation, data center construction, and model hosting to ensure total infrastructure control. Geopolitical instability and US legislative threats, such as the Patriot Act and potential tariffs, are driving European governments to adopt this "Sovereign First" model to secure data compliance and economic growth through localized "AI Growth Zones." By coordinating renewable energy surpluses with high-density AI hardware requirements, this strategy aims to retain domestic research talent and bypass the physical and legal vulnerabilities of reliance on foreign cloud providers.

A fireside Chat with Ramin Hasani, Co-Founder & CEO of Liquid AI ft. Robert Cahill (Enhance Venture)

Ramin Hasani, Robert Cahill

Liquid AI, an MIT spin-off founded by Ramin Hassani, is deploying a unique full-stack foundation model platform that utilizes biologically inspired Liquid Neural Networks to run generative AI directly on edge devices. This architecture achieves linear scaling rather than the quadratic costs of standard Transformers, enabling powerful on-device intelligence in resource-constrained hardware like smartphones and satellites without relying on data centers. Positioned as a specialized "gatekeeper" alongside massive cloud-based Oracles, the Boston-based company targets a projected $1 trillion on-device AI market by 2035 with applications spanning consumer electronics, automotive, and aerospace sectors.

SQORUS & Oracle: How AI is transforming step by step our industries. Use cases by ONET

Cyril Grira, Geoffrey, Gaurit Champenois, Romain Guyettremotier

Oracle, Skorus, and ONET Airport Services unveiled a comprehensive digital transformation at a 40th-anniversary celebration that leverages Oracle Cloud Infrastructure to reduce annual baggage losses by 25%. Led by Cyril Grira of Oracle, Geoffrey of Skorus, and ONET CIO Guillaume de Montessu, the project utilizes Autonomous Database 23ai, AI-powered vision services, and generative AI to shift operations from rigid timing-based SLAs to value-driven, real-time decision-making. Future initiatives plan to advance this system toward autonomous mission assignment and augmented reality interfaces to further optimize ground service productivity across Charles de Gaulle and other major hubs.

Fractile AI, Rogo, Scribble VC, San Francisco Compute: The AI Trillion-Dollar Opportunity

Eric Manisse, Elizabeth Weill, Walter Goodwin, John Willett

San Francisco Compute CEO Eric Manisse, Scribble Ventures' Elizabeth Weill, and founders Walter Goodwin and John Willett convened to analyze strategies for scaling AI infrastructure, hardware, and enterprise solutions amidst rapid technological evolution. The panel detailed critical operational frameworks, including balancing hardware development timelines with software agility, mitigating hallucination risks in financial applications, and proactively shaping export control regulations rather than merely complying with them. Experts concluded that long-term success in this sector relies on specialized human expertise for deep tech, the creation of durable data moats, and a shift toward treating global compute as a utility to drive the next industrial revolution.

Data Overload into Actionable Gold: AI’s Rescue Mission for Enterprises Tackling Unstructured Data

Nassime Reyholt, Victoria, Samuel Cohen, Harry Folliter, Robert, Alonzo Garcia

Representatives from Doctolib, Databricks, Alorica, Fergen, and Blackbox AI convened to discuss the critical transition from structured legacy systems to leveraging unstructured data for enterprise decision-making. The panel emphasized that successful AI deployment requires strict governance, real-time evaluation frameworks, and the use of synthetic data to ensure agents can reliably execute tasks without the 80% pilot failure rate common in the industry. By prioritizing data intelligence over model architecture, these leaders outlined a roadmap for moving from experimental playground agents to autonomous systems that self-correct in live production environments.

Fireside Chat with George Robson, Partner @ Sequoia & Felix Blossier, Co-Founder & COO @ Pennylane

George Robson, Felix Blossier

Penny Lane, founded by seven entrepreneurs after the Price Match acquisition, transitioned from a service-led hybrid model to a pure software platform designed to empower SMB accountants through AI-driven co-pilot tools rather than automation. The company navigates the complex regulatory landscapes of Continental Europe, particularly France and Germany, by leveraging a business rule management system that allows non-engineers to manage compliance updates without code changes. With a recruitment strategy prioritizing curiosity and domain expertise over experience, Penny Lane aims to scale its all-in-one bookkeeping and tax solution across Europe despite the severe scarcity of local accounting talent.

Clarifai, Oracle, TensorWave, Cherry & Vaire: Sovereign AI in a Borderless Tech World

Nadia, Rodolfo, Don Barnetson, Peter Tomasik, Hakim Lumi, Matt Zehler

Panelists including Peter Tomasik, Hakim Lumi, and Don Barnetson established that physical infrastructure and talent are the primary prerequisites for Sovereign AI, warning that Europe faces a competitive deficit while the Middle East leverages energy wealth to emerge as a third global contender. The discourse highlighted significant risks regarding stranded hardware assets due to rapid obsolescence and emphasized that current enterprise adoption remains driven by pragmatic cost and performance rather than theoretical sovereignty. Strategic recommendations from the session urge diverse silicon sourcing, cohesive national policy frameworks, and a shift toward "security by design" to address the complex vulnerabilities inherent in large language model deployment.