Subvolts take AI agents need more than models; this episode explores the essential infrastructure that makes them work. Mar 12, 2026 - Developers
- Learn about the essential infrastructure and primitives for building advanced AI agents beyond basic prompting.
- Business
- Understand how the AI agent stack is evolving, impacting the development of more capable and autonomous AI applications.
This episode features a key leader from a prominent AI developer tools company discussing practical agent infrastructure.
Everything Gets Rebuilt: The New AI Agent Stack | Harrison Chase, LangChain
Audio streams from Matt Turck. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Harrison Chase, co-founder and CEO of LangChain, discusses the evolving AI agent stack on The MAD Podcast. He explains how agents are moving beyond simple prompts to handle planning, tool use, coding, and memory. The conversation focuses on the infrastructure and new primitives needed for agents to function effectively in real-world applications, shifting the frontier from the model itself to the surrounding system.
Imagine AI agents as smart assistants. This episode explains that the real magic isn't just the AI brain (the model), but the tools and systems built around it, like memory and planning abilities, that let them do complex tasks.
This episode offers a deep dive into the practical infrastructure powering AI agents, a critical area as frontier labs like OpenAI and Google Gemini push agent capabilities. It's for developers and product leaders building AI applications who need to understand the 'stack' beyond just the large language model. Chase explains key concepts like harnesses, subagents, and memory, directly relevant to current industry efforts in creating more capable and autonomous AI systems.
- HostMatt Turck
- GuestHarrison Chaseco-founder and CEO, LangChain
- Guest: Harrison Chase, co-founder and CEO of LangChain
- Show: The MAD Podcast
- Length: 47 minutes
- Core Idea: The frontier is shifting from the model itself to the stack around the model.
- Key Agent Capabilities: Planning, tool use, coding, file management, remembering over time.
Harrison Chase, co-founder and CEO of LangChain, joins the MAD Podcast to explain why everything in AI is getting rebuilt. As agents evolve from simple prompt-based systems into software that can plan, use tools, write code, manage files, and remember things over time, the real frontier is shifting from the model itself to the stack around the model. In this conversation, we go deep on harnesses, subagents, filesystems, sandboxes, observability, memory, and the new infrastructure required to make AI agents actually work in the real world.
(00:00) Intro - meet Harrison Chase
(01:32) What changed in agents over the last year
(03:57) Why coding agents are ahead
(06:26) Do models commoditize the framework layer?
(08:27) Harnesses, in plain English
(10:11) Why system prompts matter so much
(13:11) The upside — and downside — of subagents
(15:31) Why a useful agent needs a filesystem
(18:13) The core primitives of modern agents
(19:12) Skills: the new primitive
(20:19) What context compaction actually means
(23:02) How memory works in agents
(25:16) One mega-agent or many specialized agents?
(27:46) Has MCP won?
(29:38) Why agents need sandboxes
(32:35) How sandboxes help with security
(33:32) How Harrison Chase started LangChain
(37:24) LangChain vs LangGraph vs Deep Agents
(40:17) Why observability matters more for agents
(41:48) Evals, no-code, and continuous improvement
(44:41) What LangChain is building next
(45:29) Where the real moat in AI lives
- harnesses
- Harrison Chase
- systems that help manage and orchestrate AI agents.
- subagents
- Harrison Chase
- smaller AI agents that work together on a larger task.
- context compaction
- Harrison Chase
- techniques to reduce the amount of information an AI needs to process at once.
- observability
- Harrison Chase
- tools to monitor and understand how AI agents are performing.
Same guests, other shows
40 min
47 minOpenClaw: Why the Internet Isn't Built for AI Agents
WithYoko Li, Guido Appenzeller, Joel de la Garza
48 minParamount's $110 billion Warner Bros. gamble
WithRich Greenfield, Julia Alexander
1 h 05Are We Screwed If AI Works? — With Andrew Ross Sorkin
WithAndrew Ross Sorkin
Questions people ask
- What is the main shift in AI development discussed?
- The frontier is shifting from the AI model itself to the stack of infrastructure built around the model, enabling agents to perform complex tasks.
- What are some key components of the new AI agent stack?
- Key components include harnesses, subagents, filesystems, sandboxes, observability, and memory systems.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Matt Turck.