Latest Interviews
Showing 1–15 of 645 transcripts.
Clear all filtersThe Untold Story of Higgsfield | Burning $4M a Month on AI Models | CEO, Alex Mashrabov
Alex Mashrabov, Harry Stebbings
Hicksfield achieved $1 billion in annualized revenue within 18 months by pivoting from proprietary foundational models to a cost-efficient strategy leveraging open-weights models and advanced model routing. Operating with a globally distributed workforce of 400 primarily based in Central Asia, the company drives growth through AI-native distribution for consumer entertainment and B2B workflows, recording a 300% net revenue retention rate and operating margins exceeding 80% on open-source solutions. Founder Alex now aims to scale the organization to $10 billion in revenue over the next 12 months, positioning Hicksfield as critical infrastructure for AI-generated content while rejecting traditional management theories in favor of a high-impact, meritocratic culture.
The Ads Business Model Will Die & Lessons from Working with Elon at Twitter | Parag Agrawal
Elon, Parag Agrawal, Harry Stebbings
Parallel is redefining web infrastructure as a specialized search layer optimized for agents, which the company predicts will utilize the internet 1,000 times more frequently than humans. By compressing compute costs and dynamically routing queries between low-latency voice modes and deep-reasoning background agents, the platform addresses the technical constraints of agentic workflows while disrupting traditional advertising models through data monetization at inference time. This strategic pivot positions Parallel to capture a projected $10 billion to $50 billion market as the industry shifts from human-centric browsing to event-driven, high-volume agent interactions.
Meta's Muse Hits #1 | Menlo Sounds the AI Bubble Alarm | Keith Rabois vs Airwallex: Who is Right?
Keith Rabois, Jason Lemkin, Rory O'Driscoll, Harry Stebbings
Anthropic's strategic IPO postponement to November allows for audited financial transparency, contrasting with OpenAI's massive projected capital requirements and Meta's rapid $100 billion market gain from its "Muse" agent launch. Meanwhile, new developments like Vercel's cost-efficient "Jeff" model and Meta's entry into agentic commerce are reshaping market dynamics by challenging legacy infrastructure and forcing retailers like Amazon to choose between blocking or integrating autonomous shopping agents. These technological shifts occur alongside a venture capital environment characterized by larger check sizes and selective investment decisions that prioritize data sovereignty and margin of safety over speculative valuations.
Daniel Dines on Why Work Processes Not Models Will Be The Most Valuable Asset in AI
UiPath CEO Daniel Dines argues that enterprise AI success depends on creating detailed "maps of work" to document proprietary workflows, as AI lacks the human capacity for on-the-job transformation and real-time reasoning. He outlines a strategy where organizations deploy deterministic automation generated by AI coding agents rather than relying on probabilistic models for direct execution, while simultaneously retraining employees to fill roles requiring cultural nuance and initiative. Dines projects UiPath will grow from $1.6 billion to a $50 billion valuation by evolving into a comprehensive orchestration platform, though he warns that a future with autonomous "Einstein" agents could render current approaches obsolete.
How Many Will Actually Get Built & Is Energy AI's BIGGEST Bottleneck? | Positron AI Co-founder
Thomas Sohmers, Harry Stebbings
Positron AI secured $875 million in Series C funding at a $5 billion valuation to address the critical memory bandwidth bottleneck hindering current AI inference infrastructure. Co-founder Thomas Somers details the company's strategy to deploy full-stack semiconductor solutions that overcome the "memory wall," while simultaneously challenging regulatory caps and geopolitical narratives that he argues threaten technological freedom and accelerate centralization. The event outlines a future where vertical integration and flat-rate pricing models drive autonomous agent adoption, positioning hardware scaling and energy infrastructure as the primary drivers of the next growth phase in the data economy.
Instinct Raising $1B at $10B & Meta Launches Muse | Miro Sells for $1.36B After a $17.5B Valuation
Jason Lemkin, Rory O'Driscoll, Harry Stebbings
The discourse on AI safety centers on Dario Amodei's controversial "Pace the Frontier" proposal, which sparked intense debate among industry leaders regarding regulatory intervention versus voluntary self-regulation while markets largely dismissed existential risks in favor of tangible cyber-vulnerability concerns. In the competitive landscape, Meta aggressively deployed its Muse assistant leveraging massive engineering resources and proprietary infrastructure, whereas European players secured a record $3 billion funding round from Samsung and ASML to establish AI sovereignty against US dominance. Simultaneously, the broader market underwent a significant correction illustrated by Bending Spoons acquiring Miro at a steep discount and the strategic struggles of legacy firms like Adobe to pivot toward AI-driven growth amid tightening liquidity conditions.
How LPs Allocate to Venture in 2026: What They Want, What They Don’t | Baylor University CIO
David Morehead, Harry Stebbings
Baylor University's $2.6 billion endowment office, led by Moorhead, is restructuring its 55% private and 45% public portfolio to prioritize capital velocity and upside capture amidst declining enrollment and rising tuition pressures. The team distinguishes itself through a mechanistic deployment strategy, conservative valuation discipline, and direct GP negotiations that favor annualized growth equity returns of 30% over traditional fund duration models. While maintaining a neutral stance on private credit, the fund actively bets on the resilience of legacy SaaS, permitted data center infrastructure, and biotech, aiming to navigate its expansion into the single-digit billion scale by maximizing the absolute dollar impact of its $2.5 million core positions.
Labour, Engineering, Social Media, GrokBots, Cybercabs: 7 Predictions for How AI Changes the World
Matteo Franceschetti, Harry Stebbings
Eight Sleep Co-founder Matteo Franceschetti details how the company replaces traditional engineering and marketing teams with AI agents to operate as a 160-person entity generating revenue per employee exceeding Apple's. The presentation outlines the firm's aggressive global expansion into markets like China via super-apps and its strategic pivot from a hardware manufacturer to a medical platform while navigating intense talent retention wars and rising AI infrastructure costs. Franceschetti concludes with predictions on the future of work, suggesting that Universal Basic Income will eventually become necessary as AI-driven efficiency drastically reduces the demand for human labor.
Jensen Huang Declares AGI Has Arrived | Tesla Launches Cybercabs | Index Pulls Out of Town
Jensen Huang, Jason Lemkin, Rory O'Driscoll, Harry Stebbings
Jensen Huang, Jason Calacanis, and other industry leaders debated the arrival of AGI and the economic potential of AI in coding versus law while discussing critical safety breaches that bypassed current guardrails. The conversation also covered market dynamics where aggressive startups like Instinct challenge regulations, alongside autonomous vehicle unveilings by Elon Musk and a wave of record-breaking valuations and M&A activity. Ultimately, the panel warned of an impending "Neolab" bubble correction, emphasizing that only companies capable of rapid evolution or those with massive corporate backing will survive the current capital crunch.
Town vs Instinct vs GrokBot | Why the AI Assistant Market Is Not a Bubble
Jean-Denis "JD" Grèze, Harry Stebbings
Town, led by founder Jean-Denis, has pivoted from a failed tax prep venture to a mainstream email and calendar AI assistant targeting a $100 billion valuation through a network effect at the agent level. With conversion rates exceeding 15% and a multi-model routing strategy to optimize costs, the company navigates competitive pressure from tech giants like Google and Apple by prioritizing enterprise workflow integration over consumer subsidies. JD anticipates that future market leaders will be defined by their ability to build emotional defensibility through autonomous agents while maintaining strict token economics to ensure sustainable revenue growth.
How to Build Your Own Data Center & Why Every Startup Should Do It
Cliff Weitzman, Harry Stebbings
Speechify founder Cliff Weitzman openly acknowledged a strategic error in initially avoiding the B2B API market, prompting the company to launch competitive products against rivals like 11 Labs while leveraging a proprietary hardware strategy to secure cost-efficient access to next-generation NVIDIA GPUs. By owning assets to lower training costs and prioritizing a specialized "Simba 3.2" model, Speechify aims to dominate the B2C voice space and compete directly with industry giants through a hybrid compute approach that decouples legacy inference hardware from high-performance training clusters. This infrastructure investment supports the company's pivot toward generalist AI agents and synthetic data generation, positioning Speechify to capitalize on a predicted shift from screen-based to voice-first human-computer interfaces over the next five years.
NVIDIA Crushes Quarter | OpenAI Cuts Off Cursor | Instinct Hits $2.5B Valuation
Jason Lemkin, Rory O'Driscoll, Harry Stebbings
NVIDIA reported record $96.2 billion quarterly revenue and announced strategic talks to acquire Hugging Face for $12.9 billion to secure its position in the open-weights inference market. Simultaneously, the broader AI ecosystem is defined by aggressive venture capital deployment into personal agents and coding tools, alongside intensifying conflicts between major players like OpenAI and Cursor over data access and intellectual property. These developments are driving a shift toward "compound startups" that must achieve 100x faster development velocities to survive, while emerging security threats from autonomous agents force organizations to rapidly upgrade defense protocols.
ClickHouse CEO: AI Margins Need to Improve | Revenue Concentration Should be a Concern
ClickHouse CEO Aaron Katz highlighted the company's $15 billion valuation and $350 million in Annual Recurring Revenue, projecting a potential IPO by December 2027 while emphasizing a strategic preference to remain private. Leveraging a highly efficient sales model and a distributed workforce, the firm has secured adoption across nearly every AI-native enterprise with a net dollar retention rate exceeding 200%. Despite acknowledging revenue durability risks from low switching costs in agentic workflows, Katz forecasts an unprecedented acceleration of the AI cycle and a balanced future between open-weight and frontier models.
Should American Enterprises Work With Open-Source Chinese Models? | Only 10% of Neo-labs survive
The speaker outlines a transformative AI market where valuations could reach $300 billion for data-centric entities and NVIDIA might hit $10 trillion, driven by a strategic divergence between open models for commodity tasks and proprietary fine-tuning for high-value intelligence. Key industry dynamics include Microsoft's infrastructure advantage, the consolidation of frontier labs facing regulatory or open-model pressures, and a fundamental shift in enterprise strategy toward hybrid human-AI development rather than full labor replacement. Organizational success will depend on hiring mission-aligned founders over traditional pedigrees and abandoning performative work cultures to build durable workflows that prevent the projected 80-90% failure rate of current "neolabs" within 18 months.
Anthropic's $30T Assumption & OpenAI Confirms IPO | Why Customer Service & Robotics are Overinflated
Jason Lemkin, Rory O'Driscoll, Harry Stebbings
NVIDIA's $6 billion acquisition of Poolside and subsequent strategic investments in data providers like McCor illustrate a market shift where hyperscalers finance ecosystem growth to secure chip demand, often forcing early exits for startups unable to match the capital intensity required for frontier models. Concurrently, OpenAI's scheduled 2027 public offering aims to stabilize revenue projections against Anthropic while enterprises grapple with escalating token costs that threaten to outpace traditional hiring budgets by 2027. Despite high valuations for private giants like Stripe and Databricks, skepticism remains regarding the longevity of pure-play customer support software and general-purpose robotics, as the industry increasingly concentrates wealth in Silicon Valley while leaving broader demographics behind.