Subvolts take Agent development prioritizes context engineering over just better models for long-term autonomy. Jan 21, 2026 - Developers
- Learn about advanced agent architectures, context engineering, and memory systems for building autonomous AI.
- Business
- Understand how AI agents are evolving for complex, long-term tasks, impacting future software development.
This interview provides a deep dive into the engineering behind advanced AI agents from a leading practitioner.
Context Engineering Our Way to Long-Horizon Agents: LangChain’s Harrison Chase
Audio streams from Sequoia Capital. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Sequoia Capital's Training Data podcast features Harrison Chase, cofounder of LangChain. They discuss the evolution of AI agents, focusing on long-horizon agents capable of extended autonomous work. Chase explains how context engineering, not just model improvements, drives agent development. The conversation covers coding agents, file system roles, and differences from traditional software development, including traces and self-improving memory systems.
This episode explains how AI 'agents' are being built to work on tasks for a long time without constant human help. It focuses on how engineers manage the information agents need, like giving them a good memory, to make them smarter and more useful for complex jobs.
This episode offers insight into the practical engineering challenges of building advanced AI agents. Listeners will learn how context engineering and memory systems enable agents to perform complex, long-term tasks. It's for AI developers and researchers interested in agent architectures and the future of autonomous systems. This work directly informs frontier labs like Google Gemini and OpenAI as they develop more capable, persistent AI agents for various applications, including coding assistance and complex problem-solving.
- HostSequoia Capital
- GuestHarrison Chasecofounder, LangChain
- Guest: Harrison Chase, cofounder of LangChain
- Title: Context Engineering Our Way to Long-Horizon Agents: LangChain’s Harrison Chase
- Show: Training Data (an official AI show)
- Length: 40 minutes
- Key Concept: Context engineering is fundamental to agent development.
- Agent Evolution: Evolution from scaffolding to harness-based architectures.
Harrison Chase, cofounder of LangChain and pioneer of AI agent frameworks, discusses the emergence of long-horizon agents that can work autonomously for extended periods.
Harrison breaks down the evolution from early scaffolding approaches to today's harness-based architectures, explaining why context engineering - not just better models - has become fundamental to agent development.
He shares insights on why coding agents are leading the way, the role of file systems in agent workflows, and how building agents differs from traditional software development - from the importance of traces as the new source of truth to memory systems that enable agents to improve themselves over time.
Hosted by Sonya Huang and Pat Grady
- long-horizon agents
- Harrison Chase
- AI systems that can work on tasks for a long time without needing constant human input.
- context engineering
- Harrison Chase
- Designing how AI agents manage and use the information they need to perform tasks.
- traces
- Harrison Chase
- Records of an AI agent's actions and decisions, serving as a source of truth for debugging and improvement.
1 h 08David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
WithDavid Silver, Lasse Espeholt
1 h 041 h 0552 minMaking Cities Awesome: Peregrine’s Nick Noone & Ben Rudolph
WithNick Noone, Ben Rudolph
Same guests, other shows
47 min
1 h 42Davos 2026: The US-China AI Race, GPU Diplomacy, and Robots Walking the Streets | #225
WithSalim Ismail, Dave Blundin, Dr. Alexander Wissner-Gross
1 h 51Claude Code Ends SaaS, the Gemini + Siri Partnership, and Math Finally Solves AI | #224
WithSalim Ismail, Dave Blundin, Alexander Wissner-Gross
28 minMartin Casado on the Demand Forces Behind AI
WithMartin Casado
Questions people ask
- What are long-horizon agents?
- Long-horizon agents are AI systems designed to work autonomously for extended periods on tasks.
- What is context engineering in AI agents?
- Context engineering is fundamental to agent development, focusing on how agents manage and use information to perform tasks, rather than solely relying on model improvements.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Sequoia Capital.