Lee Robinson

We just rolled out CursorBench 4.0! It includes new tasks for how well models follow instructions, work on challenging projects over time, and is more difficult than before (so all models score lower).

We just increased the included usage of Grok models in Cursor! Demand has been increasing with the launch of Grok 4.6. We already doubled limits last month with more compute. And now we're raising limits again permanently. Enjoy!

Congratulations @SpaceX team! This was incredible to see live. The energy here at Starbase is infectious.

How could we make @Bot better? Open to any feedback or feature requests! Also... ❤️ this post for a surprise!

This is a wonderful article. I love the design details. Recommended reading!

We didn't forget about @Bot either... It's now available for all paid customers. $20/mo to change how you work with computers! We've reset weekly usage limits for all Bot customers. Try it out!

Chat → Code → @Bot​s I was skeptical at first, but I'm now Bot-pilled. It is hard to give up "the old way" of prompting agents, but I strongly believe always-running bots are the future of where all work on a computer is headed. For example, it still feels *weird* to not see the response stream back to me, or not look at model pickers, and trust in the model/harness to do compaction in these very long conversations. But once you see how simple the product UX can be, and see the results, this way of working is amazing (even with it being in early beta). Models will continue to improve exponentially in the next 3-6 months, getting better and better at proactively doing real work for you in the background. And this feels like the right interface and UX for that work. I started by using Bot for personal tasks instead of work things, but it quickly took over the majority of my tokens. Very excited about the future of Bot!

The similarities between teaching my daughters to be good humans and teaching machines to think and learn new skills continue to amaze me. And yet, watching my oldest run through our yard, laughing and experiencing life, I can't see how machine intelligence could ever compare. It's already hard to imagine life without AI, much like life without the internet. But the more time I spend talking to machines, the more I appreciate the human condition and the messiness of the real world. Being human, with all its imperfections, is awesome.

Ask your @Bot to monitor the situation, no extra cost!

This architecture is also the future of coding btw. You can think about Cursor Projects like "Grok Bot for Coding". It's the same idea, just with more visibility into subagents, context, etc if you want. It's like GarageBand vs. Logic Pro, for my music friends.

Every game is being decompiled Every math problem is being solved Everything is happening all at once Think about the progress in the past year Now imagine another year And another year

I can appreciate (and even enjoy) some AI songs, but I just can't get the same emotional connection to them as I do from art created by humans. My favorite music producers all have creative ways of making their music sound more "human". You might not even realize it's there, but you can feel the difference without it. For example: 1. Many hip-hop artists find beats made in software like Ableton or FL Studio to be too "stiff". When all of the snares or kicks are perfectly on beat, it feels robotic. Producers will intentionally make things slightly off-beat (e.g. physically playing drum pads where there is some room for human error) so the beat has more "swing". 2. We've had digital music for decades, and yet producers will still add in the crackling sound effect of a record player needle hitting the vinyl. If you really listen, you'll hear this in the back of tons of R&B and acoustic songs. 3. Even for electronic music, producers will take the raw computer generated noises and add tons of artificial distortion to make them have more "texture". Some will even take a voice note on their phone of a loud thud noise, and then tweak the EQ and filters to make that be the kick drum, instead of using a fully digital sound. 4. When you are recording real instruments, like putting a microphone up to a guitar, you can also capture other sounds in the room. You'll notice on older records there are often mistakes or strange sounds captured in the recording that they end up keeping in. Ironically these have ended up being some of the most iconic parts of songs. For example, in the song "(Sittin' On) The Dock of the Bay", the iconic whistling that Otis Redding does at the end was supposed to be a placeholder for an outro, but he unfortunately died in a plane crash days after recording it. The whistling was my favorite part before I even knew that anecdote. This same thing applies to most art, like images or movies. Think about why Christopher Nolan records all of his movies on film with these massive, expensive, difficult to use IMAX cameras. He truly believes the audience can feel the difference and goes to great lengths to preserve that artistry. So, can a robot write a symphony? Yes. Is it the same as what human artists can do? No. We may train AI on all of these patterns. AI might come up with new types of art we haven't thought of before! But humans will continue to have an ingenuity and creativity that is unique. I will enjoy the slopcore AI songs about eating KFC chicken and also continue to be an avid concert goer and movie enjoyer.