ToolingPublished Top storyAll news from that day
An agent is the harness around its model, LangChain's Harrison Chase says
At Interrupt London, in a keynote published on 30 October, Chase said LangChain is launching a self-report endpoint agents can call as they run, and that Engine has scanned over 70 million traces.
Check our sources · 7 facts from 3 sources
Video: LangChain on YouTube
Key points
- Chase says what makes an agent work is everything around the model: the harness with skills, tools, context, controls and governance (02:43).
- He says LangChain is launching a self-report tool or endpoint that an agent can call as it is running (30:45).
- Chase says Engine has scanned over 70 million traces and detected over 21,000 issues (31:50); LangChain's blog of 25 September said more than 60M traces.
What happened
Harrison Chase, a co-founder of LangChain, says what makes an agent work is not the model alone: "it's not just the model, it's everything around the model. It's the harness with skills, tools, context, controls, governance" (02:43).
He said it in his keynote at LangChain's Interrupt conference in London, which LangChain's YouTube channel published on 30 October 2026.
Chase described the future as "agents running in the cloud, working in the background, proactive, always on, multi-tenant, serving many users" (12:14), and LangChain's blog says its Managed Deep Agents product combines the Deep Agents harness with managed infrastructure.
The most concrete announcement was a self-report endpoint, which Chase called "a new thing that we're actually launching today" (30:45) and described as a "tool or endpoint that the agent can call as it's running" (30:49).
On Engine, LangChain's tool that scans traces for issues, Chase said "we've seen it be used tremendously already so far, with over 70 million traces scanned and detected over 21,000 issues" (31:50). Those are his figures: LangChain's own blog post of 25 September gave "more than 60M traces" analysed since Engine's launch in May.
Everything here is LangChain describing its own products.
Underlined sentences link to the facts we checked. Dotted words open a short explainer.
Fact 1 of 7 · from LangChain (YouTube), 30 Oct 2026
(02:43) So we're really seeing that it's not just the model, it's everything around (02:46) the model. It's the harness with skills, tools, context, controls, (02:50) governance.
Fact 3 of 7 · from LangChain (YouTube), 30 Oct 2026
(30:45) And so this is a new thing that we're actually launching today, where basically (30:49) we have this self-report kind of like tool or endpoint that the agent can call (30:54) as it's running.
What it means for you
Our viewChase places the choices that shape an agent in its harness, so skills, tools, context, controls and governance are the parts he points to, beyond the model itself. His Engine figures are LangChain's account of its own product. If you run agents in production, try the self-report endpoint on one non-critical agent and compare what the agent reports about its own problems with what your traces show.
It adds no new facts.
Your reaction
We count reactions per story and day, never who reacted. The counts help us choose what goes in the monthly issue. If you are signed in, your own page shows yours too.
We count reactions per story and day, never who reacted. They help choose what goes in the monthly issue.
Check our sources
Every sentence above is checked against these 3 sources.
1 Why Own Your Intelligence? Harrison Chase Keynote | Interrupt London
Open the source archived copy-
At 02:43 Harrison Chase says in the keynote: it's not just the model, it's everything around the model. It's the harness with skills, tools, context, controls, governance. The speaker is Chase, who gives the keynote alone, speaks of our mission at LangChain, and is named in the watch page description.
cite(02:43) So we're really seeing that it's not just the model, it's everything around (02:46) the model. It's the harness with skills, tools, context, controls, (02:50) governance.
-
At 12:14 Chase says: I think it's agents running in the cloud, working in the background, proactive, always on, multi-tenant, serving many users.
cite(12:14) I think it's agents running in the cloud, working in the background, proactive, (12:18) always on, multi-tenant, serving many users.
-
At 30:45 Chase says: this is a new thing that we're actually launching today, where basically we have this self-report kind of like tool or endpoint that the agent can call as it's running.
cite(30:45) And so this is a new thing that we're actually launching today, where basically (30:49) we have this self-report kind of like tool or endpoint that the agent can call (30:54) as it's running.
-
At 31:50 Chase says: We've seen it be used tremendously already so far, with over 70 million traces scanned and detected over 21,000 issues. He is speaking of LangSmith Engine, and the figures are his.
cite(31:47) We are very very bullish on Engine. (31:50) We've seen it be used tremendously already so far, with over 70 million (31:55) traces scanned and detected over 21,000 issues.
-
The watch page, titled Why Own Your Intelligence? Harrison Chase Keynote | Interrupt London, has an upload date of 30 October 2026 and a description that begins: Harrison Chase, Co-Founder and CEO of LangChain.
citetitle: Why Own Your Intelligence? Harrison Chase Keynote | Interrupt London; uploadDate 2026-10-30T07:12:42-07:00; description: Harrison Chase, Co-Founder and CEO of LangChain, explains what it means to own your intelligence
2 New in LangSmith: Engine v2, Managed Deep Agents, Fine-Tuning, and more
Open the source archived copy-
LangChain's blog post of 25 September 2026 says Managed Deep Agents combines the Deep Agents harness with managed infrastructure, and that since launching in May, Engine has helped engineers analyze more than 60M traces.
cite
3 About LangChain
Open the source archived copy-
LangChain's About page says LangChain started as Harrison Chase's side project in late 2022 and that he teamed up with co-founder Ankush Gola to start the company in early 2023.
cite
Every sentence above is checked against these 3 sources. Each fact has its own address you can cite.
Source 1 · LangChain (YouTube) · 30 Oct 2026
Why Own Your Intelligence? Harrison Chase Keynote | Interrupt London
At 02:43 Harrison Chase says in the keynote: it's not just the model, it's everything around the model. It's the harness with skills, tools, context, controls, governance. The speaker is Chase, who gives the keynote alone, speaks of our mission at LangChain, and is named in the watch page description.
(02:43) So we're really seeing that it's not just the model, it's everything around (02:46) the model. It's the harness with skills, tools, context, controls, (02:50) governance.
At 12:14 Chase says: I think it's agents running in the cloud, working in the background, proactive, always on, multi-tenant, serving many users.
(12:14) I think it's agents running in the cloud, working in the background, proactive, (12:18) always on, multi-tenant, serving many users.
At 30:45 Chase says: this is a new thing that we're actually launching today, where basically we have this self-report kind of like tool or endpoint that the agent can call as it's running.
(30:45) And so this is a new thing that we're actually launching today, where basically (30:49) we have this self-report kind of like tool or endpoint that the agent can call (30:54) as it's running.
Show all 5 factsShow fewer facts
At 31:50 Chase says: We've seen it be used tremendously already so far, with over 70 million traces scanned and detected over 21,000 issues. He is speaking of LangSmith Engine, and the figures are his.
(31:47) We are very very bullish on Engine. (31:50) We've seen it be used tremendously already so far, with over 70 million (31:55) traces scanned and detected over 21,000 issues.
The watch page, titled Why Own Your Intelligence? Harrison Chase Keynote | Interrupt London, has an upload date of 30 October 2026 and a description that begins: Harrison Chase, Co-Founder and CEO of LangChain.
title: Why Own Your Intelligence? Harrison Chase Keynote | Interrupt London; uploadDate 2026-10-30T07:12:42-07:00; description: Harrison Chase, Co-Founder and CEO of LangChain, explains what it means to own your intelligence
Source 2 · LangChain · 25 Sep 2026
New in LangSmith: Engine v2, Managed Deep Agents, Fine-Tuning, and more
LangChain's blog post of 25 September 2026 says Managed Deep Agents combines the Deep Agents harness with managed infrastructure, and that since launching in May, Engine has helped engineers analyze more than 60M traces.
Source 3 · LangChain
About LangChain
LangChain's About page says LangChain started as Harrison Chase's side project in late 2022 and that he teamed up with co-founder Ankush Gola to start the company in early 2023.
Topics
The morning email
On the mornings we publish: the three top stories and up to four short ones. Free.
We email you a link to confirm. An issue may include one sponsor, always labelled Sponsored · Advertisement. Our emails count opens and clicks, not who made them. Unsubscribe in one click. What we keep