The podcast by and for AI Engineers! In 2025, over 10 million readers and listeners came to Latent Space to hear about news, papers and interviews in Software 3.0. We cover Foundation Models changing every domain in Code Generation, Multimodality, AI Agents, GPU Infra and more, directly from the founders, builders, and thinkers involved in pushing the cutting edge. Striving to give you both the definitive take on the Current Thing down to the first introduction to the tech you'll be using in the next 3 months! We break news and exclusive interviews from OpenAI, Anthropic, Gemini, Meta (Soumith Chintala), Sierra (Bret Taylor), tiny (George Hotz), Databricks/MosaicML (Jon Frankle), Modular (Chris Lattner), Answer.ai (Jeremy Howard), et al. Full show notes always on https://latent.space Sponsorship and business inquiries: [email protected] www.latent.space
Notion’s approach to building AI agents centers on a "software factory" model that prioritizes iterative prototyping, developer velocity, and deep integration with the company's existing data primitives. By treating agents as autonomous systems capable of self-verification and debugging, the team has moved away from ri...
The discussion centers on Harness engineering, a methodology for AI-driven software development, with Ryan Lopopolo from OpenAI. Lopopolo details his team's experience building an internal tool using AI agents that wrote over a million lines of code with minimal human intervention. A key aspect involves inverting contr...
The discussion centers on the transformative potential of AI, particularly its current state as an "80-year overnight success" built on decades of research. It highlights the shift from pattern completion to reasoning and coding breakthroughs, with AI now capable of real-world applications. The conversation explores AI...
Moonlake's founders, Fan-yun Sun and Chris Manning, discuss their approach to building world models, emphasizing structure and reasoning over pure scale. They differentiate their work from video generation models like Sora by focusing on action-conditioned models that predict the consequences of actions over longer tim...
Mistral AI's release of Voxtral TTS, their first speech generation model, is the central focus, with Guillaume Lample and Pavan Kumar Reddy from Mistral detailing its architecture and capabilities. The model supports nine languages, is cost-effective, and uses a novel autoregressive flow matching architecture with a ne...
The podcast explores the application of AI and machine learning in accelerating the discovery of new materials, particularly in chemistry. Heather Kulik, a professor of chemical engineering at MIT, shares her work on using AI to predict and optimize material properties, such as making tougher plastics by uncovering une...
Dreamer, a platform for building and using AI agents and agentic apps, aims to make AI accessible to everyone, including non-technical users. David Singleton, former CTO of Stripe and co-founder of Dreamer, details how the platform allows users to discover, build, and utilize AI agents through a personal agent called S...
The Latent Space podcast features Felix Rieseberg from Anthropic, who discusses Claude Cowork, a user-friendly version of Cloud Code. Rieseberg explains that Cowork allows non-technical users to leverage Cloud Code for tasks like managing expenses and organizing knowledge bases. He draws parallels between Cowork and Vi...
Turbopuffer's origin story and architecture are explored with Simon Eskildsen, delving into its evolution as a search engine for unstructured data. Eskildsen details Turbopuffer's unique approach to database architecture, leveraging NVMe SSDs and object storage, and its reliance on S3's consistency for consensus. He re...
The Latent Space Podcast explores the evolving landscape of AI development, developer experience, and the future of inference. Nader Khalil and Kyle Kranen from NVIDIA discuss the acquisition of Brev, a developer tool for GPU access, and NVIDIA's broader strategy in developer experience, emphasizing the importance of u...
The podcast explores Cursor's Cloud Agents, focusing on how they enhance software development workflows through automation and collaboration. A key innovation is the agent's ability to test its own code and provide a video demo of the changes, streamlining the review process. The discussion highlights three pillars: mo...
The podcast explores the transformative impact of AI agents on enterprise workflows, focusing on the challenges and opportunities in managing and securing agent access to corporate data. It highlights the shift from humans actively engaging with files to AI agents continuously utilizing data for various tasks, such as ...
METR's Joel Becker joins the Latent Space Podcast to discuss AI model evaluation and threat research, focusing on METR's work in assessing AI capabilities and propensities. Becker addresses the "Time Horizon" chart, explaining its origins and the methodology behind task selection, emphasizing the focus on economically ...
The podcast explores the ethics and practicalities of "distillation attacks" on large language models (LLMs), where smaller models are trained on the outputs of larger, proprietary models. The discussion covers the challenges of detecting such attacks versus legitimate evaluation, noting that scale and pattern analysis...
The podcast explores the intersection of AI and material science, particularly regarding climate change and new materials discovery. Max Welling discusses his career evolution from theoretical physics to AI, driven by a desire to make a tangible impact. He introduces CuspAI, his startup focused on using AI to accelerat...
Doug O'Laughlin of SemiAnalysis joins Latent Space to discuss the transformative impact of AI, particularly Claude Code, on information work and the semiconductor industry. O'Laughlin details his journey from value investing to semiconductor analysis, highlighting the pivotal insight that Moore's Law was ending, which ...
The podcast explores the limitations of the SWE-Bench Verified benchmark in evaluating coding progress due to saturation and contamination. Mia Glaese and Olivia Watkins from OpenAI discuss the original effort to create SWE-Bench Verified, involving extensive reviews by software engineers to ensure fair problem setups ...
The Latent Space podcast explores the evolving landscape of AI investment, focusing on the blurring lines between venture and growth, and infrastructure and applications. Sarah Wang and Martin Casado from A16Z discuss the unique capital flywheel in AI, where companies raise money for compute, achieve breakthroughs, and...
Jeff Dean, Chief AI Scientist at Google, joins the Latent Space Podcast to discuss the balance between frontier AI model capabilities and the practical need for efficient, deployable models. Dean emphasizes that both high-end and affordable models are crucial, with distillation techniques bridging the gap. The conversa...
Boltz, co-founded by Gabriele Corso and Jeremy Wohlwend, aims to democratize structural prediction in biology. The discussion centers on the transformative impact of AlphaFold and AlphaFold 3 on protein structure prediction and the challenges of modeling protein interactions. CASP competitions are highlighted as benchm...
Follow this podcast in Podwise
Sign in to get AI summaries, transcripts and mind maps for any episode, including new ones.
