Jonas Nelle

In personal news, @autotabai has been acquired by @cursor_ai . @alexirobbins and I have been working on giving Cursor's agent its own computer it can use like a human developer. Excited to hear what you think!

Composer 2 is out! Cursor is an example of a new type of company, not a pure app maker and not a model provider. Our aim is to build the most useful coding agents by combining the best API models and our domain-specific models.

@alexirobbins it's a yes and kind of situation

i asked composer 2 to optimize my production QA process and pitted it against gpt-5.4 composer 2's response won (as judged by both 5.4 and opus 4.6):

Composer 2 marks the one-year anniversary of our large model training efforts. Since then, we've built an exceptionally talent-dense team of ~40 people with some of the best researchers and engineers from the labs, academia, industry, and more heterogeneous backgrounds. And we are exclusively focused on coding. We don't care about models that can respond to emails, do your tax returns, or be your friend. Every FLOP, token, parameter, and researcher is entirely dedicated to software engineering.

New point on the frontier. Composer 2 has become my daily driver

i just found out that you can draw on Cursor screenshots before you send them to an agent 👀

Matt Maher tested frontier models in Cursor v. other harnesses. Cursor boosted model performance by 11% on average: Gemini: 52% → 57% GPT-5.4: 82% → 88% Opus: 77% → 93% His benchmark measures how well models implement a 100-feature PRD. @cursor_ai consistently outperformed.

gpt-5.4 in codex is often unreliable for long-running tasks. even with clear guidance, it often stops early and doesn't fully finish a task. In many cases, that only becomes obvious during code review. It ticks the tasks but there are a lot of leftovers. compared same & similar tasks in Cursor and its harness clearly performs better here.

Whoever at @cursor_ai did this. I see you. I thank you.

Coding is first, but other domains will follow quickly. Congrats @MaikTWehmeyer @maximilianeber !

@owenconti @leerob Hey, you can disable testing here #team-optional-capabilities" target="_blank" rel="nofollow noopener">cursor.com/dashboard?tab=…. Will dm, would love to hear more about what didn't go well

@leerob Hey Lee, is it possible to disable the computer usage in Cloud Agents? We've tried it since launch day but hasn't lived up to what we were expecting and is potentially costing us a ton for zero benefit.

Cursor just got a major upgrade! Agents can onboard to your codebase, use a cloud computer to make changes, and send you a video demo of their finished work. The latency of using the remote desktop is smooooth.

I noticed ⌘+F to search wasn't working on collapsed accordions. Told Cursor to fix it, went and made some breakfast, and then came back to this nice demo video. So many design details: wallpapers, chapter markers, showing keyboard input, pans/zooms, speeding up slow parts.

@TalMahaj @cursor_ai Should be enabled now!

pretty please with sugar on top @cursor_ai @jonas_nelle

🆕 Cursor's Third Era: Cloud Agents latent.space/p/cursor-third… "Cursor is no longer primarily about writing code. It is about helping developers build the factory that creates their software." — @mntruell We chat with @sjwhitmore and @jonas_nelle, both ex founders who are behind the Cloud Agents Computer launch last week, as Cursor enters its Third Era. We dive into all the technical discussions behind the tech choices, stuff that was *not* yet shipped, and @wilsonzlin's mad science experiments that have manifested in the new "Grind mode", and point the way for massively parallel, long horizon, highly autonomous agents. thanks so much to @edwinarbus for helping us get this episode together on short notice!

@TweetsOfSumit @linear Will fix this! Do you always want to use the same repo, or do you want it to pick the correct repo intelligently. The former you can configure in settings, the latter we can get out the solution for soon (it's already live in Slack)

The most frustrating thing of using Cursor Cloud Agents through @linear - by far - is the CONSTANT problem if it using the correct repo. How are others doing it? We waste so many attempts for every issue and an existing agent is not able to switch repos so we basically start from scratch. The time it takes to do that, we can just launch and prompt the agent locally.

Cursor is rolling out a new kind of agentic coding tool techcrunch.com/2026/03/05/cur…

Build the system that builds your software (and maintains it)

@snowmaker I think you might be surprised by cursor.com/onboard.

✨New demo: what if vibe coding felt more visual? @brian_lovin @maryrosecook and I did a game jam using Notion as our "IDE": launching Cursor agents from a task board, and making a custom image for each task 😎 The demo shows 3 ideas for the future of agents: 1) Agents should collaborate across apps. Each app has its focus--Notion AI is good at drafting specs and organizing tasks; Cursor is good at coding. So let them specialize! Today we're launching a new integration where Notion AI can kick off Cursor Cloud Agents to do coding tasks. The Cursor API accepts natural language prompts, so I think of this as "cross-app sub-agents" -- it's kinda cute how it resembles humans hiring outside contractors 😊 BTW: the parallelism of cloud agents is incredibly freeing for creativity, but it also creates a new problem: sooo much work to keep track of! Which brings us to the next idea... 2) Agent orchestration is a data visualization problem. A powerful frame for designing agent UIs is to think of the chat transcripts as the "raw data" and ask: what visual projections might help people make sense of this data at scale? We need to engage our human GPUs -- our visual processing -- to understand what the computer GPUs are doing for us! One thing we can do is use AI to populate traditional UIs like progress bars and status updates. But there are also new possibilities now... For example: when you have a lot going on, it can be hard to identify tasks just by text titles. So we tried generating an AI image for each task -- turns out this helps a lot by giving it a unique visual identity! And of course, it also just makes it super fun to build with friends 😃 Speaking of friends... 3) The future of coding is collaborative. Sometimes it feels like IC engineers are being reduced to middle managers: shuffling information between the team's context and the coding agents that they individually manage. The solution: bring all the people and agents into one shared space, with shared context and visibility! In the video you can get a glimpse of how this feels. Mary, Brian and I record ourselves chatting about ideas, and then we use AI to turn that conversation into a list of tasks on a shared board. As the ideas get built in parallel, we can all monitor progress and review the work together, nothing is siloed. My main takeaway from this game jam was: damn, creativity with friends, at the speed of conversation, is incredibly fun. --- Our goal here is to let anyone use Notion as a fun and creative "software factory" to build software together with your team. Give the Cursor integration a shot and let us know what you think! (AI Image gen in Notion isn't GA yet, but coming soon and already out to some users) And let me know if you'd want a template or more detailed instructions on the setup we showed in this demo...