AI + a16z
Technology
About
Artificial intelligence is changing everything from art to enterprise IT, and a16z is watching all of it with a close eye. This podcast features discussions with leading AI engineers, founders, and experts, as well as our general partners, about where the technology and industry are heading.
Episodes
- From Vector Databases to Knowledge Engines: The Next Layer of AI
Peter Levine and Pinecone CEO Ash Ashutosh discuss the evolution from vector databases to knowledge engines and how AI systems need to adapt for machine-to-machine interactions. They explore the inefficiencies of current retrieval systems…
- Why We Need Continual Learning
Malika Aubakirova and Elena Burger discuss the challenges of continual learning in AI systems. They explore the limitations of in-context learning and the need for AI models to evolve to learn from experience.
- The Agent Era: Building Software Beyond Chat with Box CEO Aaron Levie
Aaron Levie, CEO of Box, joins Erik Torenberg, Steve Sinofsky, and Martin Casado to discuss the agent era and its effects on enterprise software. They explore why coding agents succeed, the implications of abstraction layers for the workfo…
- Rethinking Git for the Age of Coding Agents with GitHub Cofounder Scott Chacon
Matt Bornstein interviews GitHub cofounder Scott Chacon about the history and future of Git. They explore how the GitButler platform is evolving version control to better support human developers and AI coding agents.
- How AI Is Reshaping IT Services from the Inside
Joe Schmidt interviews Treeline CEO Peter Doyle about the integration of AI and automation within the IT managed services sector. They discuss industry growth strategies, the challenges of software in service models, and the role of forwar…
- Patrick Collison on Stripe’s Early Choices, Smalltalk, and What Comes After Coding
Patrick Collison, CEO of Stripe, discusses his background with Smalltalk and Lisp, Stripe's early technical decisions, API design, and the impact of AI on economic productivity with Michael Truell, CEO of Cursor.
- OpenClaw: Why the Internet Isn't Built for AI Agents
Yoko Li, Guido Appenzeller, and Joel de la Garza discuss the challenges of integrating AI agents with current internet infrastructure, focusing on OpenClaw. They explore issues regarding identity, permissions, security, and the lack of inc…
- What's Missing Between LLMs and AGI - Vishal Misra & Martin Casado
Vishal Misra joins Martin Casado to discuss the mechanisms behind LLMs, covering experimental research on transformer predictions and the requirements for achieving AGI, including continuous learning and causal reasoning.
- Replit's CEO on Vibe Coding, Wealth Building, and What Most People Get Wrong About AI
Replit CEO Amjad Masad joins host Jack Neel to discuss how AI empowers non-technical users to build software. The conversation covers Replit’s growth, business milestones like rejecting a $1 billion acquisition offer, and Masad’s perspecti…
- Jack Altman & Martin Casado on the Future of VC
Jack Altman and Martin Casado of a16z discuss the changing venture capital landscape, the importance of media, and the rise of AI infrastructure. They also cover firm specialization and the competition for talent in the current market.
- AI’s Capital Flywheel: Models, Money, and the Future of Power
Martin Casado and Sarah Wang from a16z join the Latent Space podcast to discuss the unique characteristics of the current AI investment cycle. They explore the blurring lines in venture capital, the dynamics of model funding, and the gap b…
- Durable Execution and the Infrastructure Powering AI Agents
Raghu Raghuram and Sarah Wang from a16z interview Temporal CEO Samar Abbas about the role of durable execution in infrastructure for AI agents. The discussion covers state management, recoverability, and the challenges of scaling backgroun…
- Evals, Feedback Loops, and the Engineering That Makes AI Work
Martin Casado and Braintrust CEO Ankur Goyal discuss engineering in AI, including open versus closed source models, international model development, and potential market saturation. They also examine the role of benchmarks in evaluating AI…
- Sam Altman on Sora, Energy, and Building an AI Empire
Ben Horowitz, Erik Torenberg, and OpenAI's Sam Altman discuss the growth of OpenAI, the development and release of Sora, internal model usage, and the future of AI.
- Why This Isn't the Dot-Com Bubble | Martin Casado on WSJ's BOLD NAMES
Martin Casado of a16z joins WSJ’s Bold Names to discuss why current AI spending differs from the dot-com bubble. He explores how tech giants' financial strength and potential in the long tail of AI companies distinguish the current market…
- Martin Casado on the Demand Forces Behind AI
a16z General Partner Martin Casado discusses the impact of AI on infrastructure, software development, and enterprise purchasing. He explores how regulatory constraints on power and compute influence industry growth and the role of agent-d…
- How Mintlify Is Rebuilding Documentation for Coding Agents
Mintlify cofounders Han Wang and Hahnbee Lee discuss how coding agents are transforming developer documentation into infrastructure for AI tools. They also cover the company's early growth, the significance of self-healing documentation, a…
- Inferact: Building the Infrastructure That Runs Modern AI
The founders of Inferact discuss the challenges of AI infrastructure, the development of the open-source project vLLM, and their goal to create a universal inference layer for running AI models across diverse hardware environments.
- How Should AI Be Regulated? Use vs. Development
Martin Casado joins Jai Ramaswamy and Matt Perault to discuss AI regulation, focusing on the distinction between AI use and development. The conversation covers marginal risk, the importance of open source, and leveraging past technology p…
- Michael Truell: How Cursor Builds at the Speed of AI
Cursor CEO Michael Truell joins a16z's Martin Casado to discuss the development of the code editor. They explore the company's strategic focus on power users, their hiring process, and the decision to build their own editor.
- Dylan Patel on the AI Chip Race - NVIDIA, Intel & the US Government
Dylan Patel, Sarah Wang, and Guido Appenzeller discuss the implications of Nvidia's investment in Intel. The panel examines the impact on the semiconductor industry, AI infrastructure, and the global chip market.
- Feed Drop from The Generalist: Why a16z's Martin Casado believes the AI boom still has years to run
a16z General Partner Martin Casado joins The Generalist Podcast to discuss his outlook on the AI market cycle, the potential of AI coding, the evolution of a16z, and the role of open-source models. He also shares his perspective on why foc…
- Fei-Fei Li: World Models and the Multiverse
Fei-Fei Li, Erik Torenberg, and Martin Casado discuss the development of world models, exploring the potential of AI to understand and reason about the 3D physical world beyond language-based systems. They examine the role of spatial intel…
- Building the “See Something, Say Something” AI for Every Camera
Martin Casado of a16z speaks with Ambient CEO Shikhar Shrestha about the development of agentic AI for physical security. They discuss the company's Pulsar VLM and their process for creating systems that monitor camera feeds to detect and…
- The AI That Found A Bug In The World’s Most Audited Code
Matt Knight, head of the AI agent Aardvark at OpenAI, joins Joel de la Garza to discuss how AI models are evolving to identify complex security vulnerabilities in code. They examine the capabilities of current models and the potential for…
- The Death of Data Gatekeeping: AI Makes Everyone An Analyst | Hex Cofounder
Barry McCardel, CEO of Hex, talks with Sarah Wang of a16z about how AI is changing data analysis, moving past traditional dashboards. They explore agent workflows, conversational interfaces, and how Hex aims to empower everyone to work wit…
- Why Social Engineering Now Works on Machines
Ian Webster created PromptFoo to address AI security vulnerabilities, inspired by observing users bypass AI agents. His tools use conversational methods to find weaknesses, preventing security incidents before they impact customers.
- “Anyone Can Code Now” - Netlify CEO Talks AI Agents
Netlify CEO Matt Biilmann discusses the unexpected surge in users, driven by AI tools enabling non-developers to build applications. He explains how this shift expands the market for developer tools and is reshaping software economics.
- From Code Search to AI Agents: Inside Sourcegraph's Transformation with CTO Beyang Liu
Sourcegraph CTO Beyang Liu shares that 90% of his company's code now comes from AI agents. He also raises concerns that competitive open-source coding models predominantly trace back to Chinese labs, potentially impacting the US AI future.
- Ryo Lu (Cursor): AI Turns Designers to Developers
Ryo Lu, Head of Design at Cursor, shares how the tool allows designers to code at the speed of thought. The discussion covers AI's role in design, the nature of purposeful applications, and innovative interface concepts like System 7.
- How Foundation Models Evolved: A PhD Journey Through AI's Breakthrough Era
Omar Khattab, a Stanford PhD, discusses his journey creating DSPy, a new programming paradigm for LLMs that addresses the gap between user intent and AI capabilities. He argues for an interface between code and natural language to make AI…
- TruffleHog Creator: You Can’t Have AI Agents Without Secrets
Dylan Ayrey, cofounder and CEO of Truffle Security, joins Joel de la Garza to discuss the security stack for AI agents. The conversation covers the importance of protecting secrets for reliable AI agents and how Truffle Security addresses…
- Tigris Data CEO on Building Your Own Datacenters
Martin Casado of a16z talks with Ovais Tariq, CEO of Tigris Data, about the difficulties of independent storage, the realities of running datacenters, and predictions for the future of cloud computing.
- Pylon: Reimagining B2B Customer Support
Pylon's cofounders share insights into their B2B customer support platform, their competitive strategy, and the demanding work culture behind their startup's success.
- Keycard: 2026 is the Year of Agents
In 2025, AI agents began to appear, and 2026 is predicted to be the year companies integrate them into production. This conversation explores the evolution from copilots to agents, the security aspects of tool-calling, and strategies for m…
- The Trillion Dollar AI Software Development Stack
a16z partners Guido Appenzeller and Yoko Li discuss the emerging trillion-dollar market for AI in software development, focusing on how "agents with environments" are transforming the development loop and identifying founder opportunities…
- Material Security CEO: How To Find Your Ideal Customer
Abhishek Agrawal, CEO of Material Security, discusses with a16z General Partner Martin Casado how to discover an ideal customer profile and manage customers. They also touch upon how frothy markets can distort real signal.
- When Deepfakes Steal $30M: The New Edge of Cybercrime
AI is changing cybersecurity, with attackers using LLMs for fraud and defenders automating detection. This episode features Kevin Tian, CEO of Doppel Security, discussing AI
- How AI Will Change Photography Forever
AI is revolutionizing photography, moving beyond filters to AI-powered retakes that preserve identity and context. This evolution traces from early light-field cameras to current generative AI models, fundamentally changing how we capture…
- How OpenAI Built Its Coding Agent
General Partner Anjney Midha interviews Alexander Embiricos, a product lead at OpenAI, about the coding agent Codex. They discuss its origin story, its high success rate in shipping pull requests, the safety considerations for autonomous a…
- Models, Modalities, and Memes: Creating Compelling AI Characters
Hedra founder Michael Lingelbach joins AI + a16z to discuss building AI-native video focused on characters. They explore Hedra's real-time video models for expressive, dialogue-centric content, its application in meme creation and enterpri…
- Performance and Passion: Fal's Approach to AI Inference
Fal cofounders Burkay Gur and Batuhan Taskaya discuss their generative media cloud, an inference platform optimized for speed and performance. They cover infrastructure, sales, hiring, and emerging trends in AI modalities, particularly vid…
- How to Vibe Code Securely
Feross Aboukhadijeh shares insights on how Large Language Models are changing software engineering workflows. He stresses that AI-assisted programming requires the same security diligence as traditional coding, including vetting packages a…
- AI Is Upending SaaS Pricing
Martin Casado and Scott Woody discuss how artificial intelligence is fundamentally changing the value proposition of software, leading to a shift in SaaS pricing models from user-based to output-based monetization. They explore emerging pr…
- The AI Infrastructure Stack with Jennifer Li
Jennifer Li explores how AI is transforming the software stack, creating a need for new middleware and infrastructure. She discusses emerging categories and the potential for infrastructure providers to power future intelligent application…
- AI's Unsung Hero: Data Labeling and Expert Evals
Labelbox CEO Manu Sharma and a16z Infra partner Matt Bornstein explore the evolution of AI data labeling and evaluation, focusing on the shift to foundation models and generative AI. Sharma details Labelbox's adaptation by creating a netwo…
- AI, Data Engineering, and the Modern Data Stack
dbt Labs CEO Tristan Handy joins a16z's Jennifer Li and Matt Bornstein to discuss the evolution of data engineering, the impact of AI on analytics, and the importance of human verification in AI-driven workflows.
- Enabling Agents and Battling Bots on an AI-Centric Web
Arcjet CEO David Mytton and a16z partner Joel de la Garza discuss the complexities of managing web access in an AI-centric environment. They explore challenges in distinguishing between malicious bots and AI agents, and the importance of r…
- Giving New Life to Unstructured Data with LLMs and Agents
Anant Bhardwaj and Guido Appenzeller explore the impact of Large Language Models (LLMs) and AI agents on analyzing unstructured data. They cover Instabase's approach to extracting insights from complex documents and the future of AI agents…
- Beyond Leaderboards: LMArena’s Mission to Make AI Reliable
LMArena cofounders Anastasios N. Angelopoulos, Wei-Lin Chiang, and Ion Stoica discuss AI evaluation with Anjney Midha. They explore LMArena's approach to testing AI models with users for reliability, moving beyond traditional benchmarks, a…