Show: Training Data · Publisher: Sequoia Capital · Host: Sonya Huang, Pat Grady
Episode URL: https://pscrb.fm/rss/p/traffic.megaphone.fm/CPUAI9761150620.mp3
Publish date: 2026-01-21
Duration: NAs
Default source credibility: HIGH — Sequoia partners interview frontier-lab founders + F500 AI buyers. VC-hosted — portfolio-company framing on recommendations; named guest metrics stay HIGH. Peer-tier to No Priors in quality.
- Long horizon agents are becoming more reliable and useful, especially in coding and research tasks, due to better models and harnesses.
- Traces and memory are crucial for understanding and improving agent behavior, enabling iterative development and human judgment.
- The development of long horizon agents differs significantly from traditional software development, emphasizing the importance of context engineering and asynchronous management.
Extracted quotes
| # | Credibility | Speaker | Org | Timestamp | Topic | Quote |
|---|---|---|---|---|---|---|
| 1 | HIGH | Harrison Chase (Founder) | LangChain | 02:09 | 01-ai-native-landscape | I think the idea of running an LLM in a loop and just having it go was always the idea of agents from the start. The issue is the models weren’t really good enough and the scaffolding and harnesses around them weren’t really good enough. And I think the models got better. We learned more about what makes a good harness over the past few years. And now they start to like really, really work. |
| 2 | HIGH | Harrison Chase (Founder) | LangChain | 18:07 | 01-ai-native-landscape | So like writing scripts is like really useful for that. And I think a coding agent can be general purpose, but I don’t know if that means that today’s coding agents are, if that makes sense, because I think a lot of the coding agents today are pretty optimized for coding tasks. |
| 3 | HIGH | Harrison Chase (Founder) | LangChain | 19:11 | 02-corporate-tools | And so what this means is that you can’t just look at the code and tell exactly what the agent would do in a specific scenario. You actually have to run it. And so what does that mean? I think like one thing that that means is that in order to tell what application is actually doing, you can’t look at the code. You have to look at actually what it does in real life. |
Per-quote detail
1. Harrison Chase — LangChain (02:09)
I think the idea of running an LLM in a loop and just having it go was always the idea of agents from the start. The issue is the models weren’t really good enough and the scaffolding and harnesses around them weren’t really good enough. And I think the models got better. We learned more about what makes a good harness over the past few years. And now they start to like really, really work.
- Credibility: HIGH — Named exec at identifiable org with specific claim about model and harness improvements.
- Topic tag:
01-ai-native-landscape
2. Harrison Chase — LangChain (18:07)
So like writing scripts is like really useful for that. And I think a coding agent can be general purpose, but I don’t know if that means that today’s coding agents are, if that makes sense, because I think a lot of the coding agents today are pretty optimized for coding tasks.
- Credibility: HIGH — Named exec at identifiable org with specific claim about coding agents.
- Topic tag:
01-ai-native-landscape
3. Harrison Chase — LangChain (19:11)
And so what this means is that you can’t just look at the code and tell exactly what the agent would do in a specific scenario. You actually have to run it. And so what does that mean? I think like one thing that that means is that in order to tell what application is actually doing, you can’t look at the code. You have to look at actually what it does in real life.
- Credibility: HIGH — Named exec at identifiable org with specific claim about agent development.
- Topic tag:
02-corporate-tools
Extracted 2026-04-14T17:23:42 via scripts/podcast_mine.py (MLX mlx-community/Qwen2.5-32B-Instruct-4bit).