scripod.com
Cursor's Third Era: Cloud Agents

Overview

Shownote

Highlights

Transcript

Chapters

Pins

Cursor's Third Era: Cloud Agents

Latent Space: The AI Engineer Podcast

Mar 06
Latent Space: The AI Engineer Podcast

Latent Space: The AI Engineer Podcast

Mar 06
OverviewShownoteHighlightsTranscriptChaptersPins

Shownote

All speakers are announced at AIE EU, schedule coming soon. Join us there or in Miami with the renowned organizers of React Miami! Singapore CFP also open! We’ve called this out a few times over in AINews, but the overwhelming consensus in the Valley is t...

Highlights

This episode dives into Cursor's latest cloud agents launch—a major evolution in AI-powered software development that moves far beyond autocomplete and local coding assistants.
00:00
Using models from different providers had strengths, resulting in a synergistic output
00:53
Cloud agents now give the model full computer use, enabling driving new features
02:08
The agent returns a tested PR ready for review, similar to what a human should do when asking for a review
02:57
Testing is enabled by default for PRs, but can be skipped with /no test or customized in agents.md
03:35
Demo videos are an easy entry point for code review and help with alignment, especially when requests are underspecified
06:10
Around Claude 4.5, full pixel-based automation became feasible
10:49
/repro automatically reproduces bugs, records videos of the process, and fixes them
32:02
The web UI aims to keep it minimal and expose shared app surfaces for both users and agents
32:39
Removing the Files app encourages delegation to agents instead of manual file handling
34:34
The big bottleneck is getting code out, not deployment infrastructure
36:34
Every agent lab should have a router to decide the best model for users
40:54
Andrej Karpathy called it the council, and there's an internal command '/council' written by Ian
1:03:35
Agents must be self-auditable and self-aware to understand their environment, secrets, and system constraints

Chapters

Agentic Code Experiments
00:00
Why Cloud Agents Matter
00:52
Live Demo Tour
02:07
Testing Defaults and Controls
02:55
Videos and Remote Control
03:35
Meta Agents and More Demos
06:09
Bug Repro Workflow
10:48
Minimal Web UI Surfaces
32:01
Why No File Editor
32:39
Full Stack Hosting Debate
34:34
Model Choice and Auto Routing
36:33
Best Of N and Model Councils
38:33
Subagents for Context
41:40

Transcript

Samantha: So this is another experiment that we ran last year and didn't decide to ship at that time, but may come back to you. LLM did, but one that was also agentic and could write code. So it wasn't just picking, but also taking the learnings from two m...