The AI Record
you can scale this to like 100 or 1,000 agents. I'm not sure how many we are running today, but I bet that's like multiple thousands a day, just agents running in the same or like copies of the code base.

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is talking about running cloud agents, each in its own virtual machine, and guesses how many Cursor itself runs; he says he does not know the number.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 21:54.

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools

Filed under

Who is speaking

Eric Zakariasson, Engineer, Developer Experience and Product, Cursor. Works inside Cursor and describes how his own team builds and runs the product with agents, so he is an insider on those practices; Cursor sells AI coding tools and the talk shows its own product, so it is promotional.

Cursor (Anysphere) sells AI or agent tools or services, so it has a reason to publish flattering numbers about AI. The passage stays in the archive; read it with that in mind.

How this was checked

Checked against the source by a first reader, then confirmed independently by a second reader who fetched the source without seeing the first reader's notes (Independent review agent (Claude Sonnet), blind second pass from passage text and citation only, no repository access, 2026-11-04).

Whisper transcript at about 21:54–22:10. The passage matches the transcript word for word, including 'I bet that's like multiple thousands a day'.

How to cite this passage

Eric Zakariasson (Cursor (Anysphere)), “Building your own software factory”, AI Engineer (YouTube), 2026-04-28, at 21:54. https://www.youtube.com/watch?v=rnDm57Py54A Archived at https://the-ai-record.com/passages/cursor-multiple-thousands-of-agents-running-a-day-2026

More from the same source

we had code owners in our code base and they were kind of right most of the time, like 80% of the time. But for these 20% of the time, they cost a lot of bottlenecks for us internally. … So what we started doing was building this agentic code owner thing. And what it essentially does is look in at PRs and check in like, first of all, what's the risk of this? What's the risk level? … And if it is a low risk, it can just approve the PR …

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is describing an internal Cursor tool that replaces some human code-owner reviews of pull requests.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 27:55.

  • 80% code owner correct rate, as described in the 2026 talk

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools Passage page

the best example of a rule that I can think of internally is for Bugbot. So when we're doing database migrations, we're not really using foreign keys on a database for performance reasons. And the models like the right way to do this is use foreign keys, right? So they will always add a foreign key. But when it … hits … and there's a PR created, we have Bugbot … it's like, oh, I have this rule saying like we should never use foreign keys. So then it flags this.

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is giving an example of a rule Cursor keeps for its Bugbot code-review tool, in an answer about writing rules for agents.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 45:06.

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools Passage page

again, cloud agents also come from this … internal need of us just wanting to … run things asynchronously. And because of that … we have also … put a lot … of effort into making our own code base work really well in cloud agents.

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is answering an audience question about cloud agents that go off on tangents, and says where Cursor's cloud agents came from.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 1:21:41.

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools Passage page

Designers work, I would say like 50/50 in Figma and code at this point … of them do code. All of them … do push to production.

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is answering an audience question about how product managers, designers and engineers work together at Cursor.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 1:18:40.

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools Passage page

And we do this for like everything. So once you post something in Slack, we either have a Linear Slack agent look at it or we have a Cursor automation to … look at the message I was posted and triage it and … look for duplicates. Or … if it's determined to be easy, … start to implement the fix for it immediately.

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is describing how Cursor's team handles messages posted in Slack, continuing an answer about connecting Linear to cloud agents.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 1:10:05.

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools Passage page

So for every ticket that's getting created in Linear, we spawn in the cloud agents. … if we have a feature flags for a specific thing that's rolled out. And if it's rolled out for two weeks with 100%, the system kind of like a single source like, hey, it's a stale feature flag at this time. You can remove it. So then we have this to create an automatic issue in Linear. And since that is hooked up with Cursor, it triggers a cloud agent to remove the feature flag. So it's kind of like completely automatic once the system knows that it's rolled out to everyone.

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

Context. Zakariasson is answering an audience question about whether agents read tasks from an issue tracker, and describing how Cursor's own team does it.

Talk: Building your own software factory, AI Engineer (YouTube), April 28, 2026, at 1:09:19.

  • 14 days feature flag full rollout days, as described in the 2026 talk

Checked against the source [ Grade B ] Said in the last year Cursor (Anysphere): $4 billion revenue, annualised, early June 2026 Sells AI tools Passage page

Something wrong? Report a correction. JSON: /api/v1/passages/cursor-multiple-thousands-of-agents-running-a-day-2026.json