Bennett Brownlow

150 posts

Bennett Brownlow banner
Bennett Brownlow

Bennett Brownlow

@bennett2b

enterprise @cursor_ai

sf Katılım Şubat 2025
181 Takip Edilen450 Takipçiler
Bennett Brownlow retweetledi
n2parko
n2parko@n2parko·
better team marketplaces are now available in @cursor_ai team marketplaces make it easy to distribute and discover skills, plugins, MCPs, subagents, hooks across your team - repo-backed or repo-less - set tools as required or discoverable - team-level configs coming soon
English
6
9
120
9K
Bennett Brownlow retweetledi
Jediah Katz
Jediah Katz@jediahkatz·
Let's talk about why Cursor's agent harness is so good. There's a misconception that first-party harnesses from the labs will always outperform. For many reasons, that isn't true. There are roughly 6 layers that go into a good agent harness: orchestration, context, routing, transport, state, and execution (tools). Some of those involve careful context engineering, and others look a lot like the traditional craft of building great software. Each of these layers needs to be optimized. And if any one of them is degraded, it can severely impact your experience with the agent. This blog post is mostly focused on context and tools, and there's still so much more to talk about there. We also want to spotlight the other (very crucial!) areas soon. There's a lot more misinformation around the agent harness out there. I'll keep writing about how we build it behind the scenes.
Jediah Katz tweet media
Cursor@cursor_ai

Our agent harness makes models inside Cursor faster, smarter, and more token-efficient. Here's how we test improvements to the harness, monitor and repair degradations, and customize it for different models. cursor.com/blog/continual…

English
31
53
857
88.7K
Bennett Brownlow
Bennett Brownlow@bennett2b·
"Bugbot is amazing, but we'd also like something optimized for security that scans new PRs and our existing codebase" -every enterprise customer CSR does both and uses the best model for the job, regardless of what lab the model comes from.
Cursor@cursor_ai

Cursor Security Review is now available for Teams and Enterprise plans. Run two types of always-on agents: 1. Security Reviewer checks every PR for vulnerabilities and leaves comments. 2. Vulnerability Scanner runs scheduled scans of your codebase and posts findings in Slack.

English
2
1
36
2.9K
Bennett Brownlow retweetledi
Cursor
Cursor@cursor_ai·
Our agent harness makes models inside Cursor faster, smarter, and more token-efficient. Here's how we test improvements to the harness, monitor and repair degradations, and customize it for different models. cursor.com/blog/continual…
English
51
101
1.3K
185.1K
Bennett Brownlow retweetledi
eric zakariasson
eric zakariasson@ericzakariasson·
cursor sdk launched yesterday! people are already putting cursor agents in places they already work: gmail, chrome, ci, terminal, docs github issues here are 11 projects built in the first day ↓
English
36
54
930
88.5K
Bennett Brownlow retweetledi
Parth Jadhav
Parth Jadhav@ParthJadhav8·
This is crazy !! Cursor now has built a Kanban board where you can just drop in tasks and the agent will pick those up and complete them.
English
176
259
5.5K
511.4K
Bennett Brownlow
Bennett Brownlow@bennett2b·
@kevinkern One of my favorite recent features! We've tried a few different modes of multi-agent management but so far Multitask feels pretty good.
English
1
0
1
187
Kevin Kern
Kevin Kern@kevinkern·
cursor shipped /multitask a bit silently, but it’s their new orchestration feature. instead of one agent doing everything in one long context, the main agent can split work into smaller async subagents. so the main chat stays clean, while side agents go off and do messy stuff like: - inspect files - plan - code - review & tests - report back with results in the main chat. - repeat orchestration and subagents aren't new. but the ux is really good, and you can use different models for different tasks within cursors harness. i've been running opus 4.7 as the main orchestrator and gpt-5.5 high for implementation tasks. this combination was the most successful, and more of my runs were working for hours.
Kevin Kern tweet media
English
12
4
127
9.5K
Bennett Brownlow
Bennett Brownlow@bennett2b·
@jediahkatz focuses on building the harnesses for each model we support in Cursor. Each model's harness is unique, and there is a lot of nuance to how they're engineered and optimized to work in Cursor. Very cool to see that GPT-5.5 in Cursor is at the top of the leaderboards.
Jediah Katz@jediahkatz

New analysis from @EndorLabs benchmarking correctness and security of coding agents. Cursor's optimized harness once again coming out on top. 👑

English
0
2
12
375
Bennett Brownlow retweetledi
Kieran Klaassen
Kieran Klaassen@kieranklaassen·
Yes, running Multitask is amazing, I can now use Opus 4.7, Composer 2, Gemini 3.1 Pro and GPT-5.5 At the SameTIMEeE?? Amazing @cursor_ai
Kieran Klaassen tweet media
English
7
6
124
23.6K
Bennett Brownlow retweetledi
ian
ian@shaoruu·
tip for @cursor_ai 3: cmd-N to create a new agent, then cmd-dot to quickly switch between projects to fire agents into @lilingxi01 is working on making everything keyboard navigable, a lot more to come🙏
English
9
3
106
5.1K
Bennett Brownlow retweetledi
lauren
lauren@poteto·
i'm increasingly convinced that the value of orchestrating many agents in parallel comes from going deep, not broad. you want to go deeper on a single or a few problems, so you can maximize your chances of getting great results: - best of N style races to find the best solution - adversarial review - multiple agents trying to repro a reported issue - use different models for different types of workloads to quote @mattpocockuk, "code is not cheap". this beats trying to context switch across many agents working on 10 different problems. the bottleneck is still me trying to remember and keep in my own context window what my agents work on. i think there will be ways to solve this with better ui. there's also the element of trust. the more trust you have, the higher up the perspective ladder you can go. when you have little trust, you must micromanage each and every agent. when you can let go, you can delegate more. this is exactly like managing people!
English
10
15
195
12.5K
Bennett Brownlow retweetledi
Lee Robinson
Lee Robinson@leerob·
Extremely excited about this Cursor ship. It's now *way* easier to run many agents at the same time (with async subagents and improved worktrees). And you can run agents across many repos (e.g. frontend and backend!)
Cursor@cursor_ai

Introducing /multitask in the new Cursor 3 interface. Cursor can now run async subagents to parallelize your requests instead of adding them to the queue. For already queued messages, you can ask Cursor to multitask on them instead of waiting for the current run to finish.

English
49
29
582
58K
Bryan Jackson
Bryan Jackson@1BJax·
@cursor_ai Sigh we use teams and jira I'll probably set this up for side projects though. 😊
English
1
0
1
422
Cursor
Cursor@cursor_ai·
Mention @​Cursor to kick off tasks in Slack and see updates of its work streaming in real time. Cursor uses context in the thread and broader channels to create a PR for you to review and ship.
English
56
67
995
136.4K
Bennett Brownlow retweetledi
edwin
edwin@edwinarbus·
One year of Cursor Pro for Lenny’s Newsletter subscribers!
Lenny Rachitsky@lennysan

25 world-class AI products, free for a full year I've partnered with @Cursor_ai, @GoogleAI, @NotionHQ, @Fin_ai, @Supabase, and @Gumloop to give my paid subscribers a free year of their tools. That's on top of the 19 premium AI products you already get free as a paid subscriber, including @Replit, @Lovable, @FactoryAI, @ManusAI, @Railway, @GammaApp, @Canva, @n8n_io, @Linear, @PostHog, and more. 25 of the hottest, most important, and most beautifully crafted products in the world, free with your Lenny’s Newsletter subscription. This all sounds way too good to be true, but it isn’t. → Already a paid subscriber? Grab your free products here: lennysproductpass.com → Not yet a subscriber? Learn more here: lennysnewsletter.com/p/new-a-free-y… Why am I doing this: There’s a growing divide—in terms of career opportunities, financial outcomes, and how much you enjoy your work—between the people who benefit from the transformation we’re living through and those who get left behind. I want to make sure you’re in the first group. Join us: lennysnewsletter.com/p/new-a-free-y…

English
33
8
176
58K
Jediah Katz
Jediah Katz@jediahkatz·
Today marks my anniversary at Cursor. When I joined, it felt like we were just a few dozen kids running around to put out ten new fires every day, hopeful that our focus and wits could transform an industry. Over the past year we've grown the company (and revenue) more than tenfold; improved our agent to work for hours or even days; launched our own frontier model, Composer; and defined a new category that the biggest players are now chasing. Here's to the next chapter!
Jediah Katz tweet media
English
29
8
378
29.6K

Keşfet

@cursor_ai @5harath @benln @kevinkern @WolfBenchAI @WolframRvnwlf @jediahkatz @lilingxi01