Why cancel compose pop!!

Not quite sure what the major difference is. Apart from UI changes and terminal inside composer. All things that "agent" is doing now was already possible before with previous cursor composer.

😂

This explanation was great, can't wait for the second one. I'm a frontend engineer and just started getting projects building assistant UI/UX so this is soooo useful to understand.

Thank you Jesus

best channel ! , I hope you definitely do a follow with details of how to manage this.. I need 'user artefacts' trying to also make UI elements dynamically draw through prompts, for my client. I'm struggling with the reasoning about how to present the elements and managing it...

What is the value here? Reading out the blog post?

Very interesting video! I suggest a deep dive into the backend of this setup for part 2, with an even more detailed analysis of the system prompt. Looking forward to seeing this evolve into a series of videos.

Youre becoming one of my favorite IDE tester out there pleasent voice and flow aswell

you are one of the best channels on all things tech 💯 up to the point and hugely informative, thanks mate for your helpful videos, cheers

Is there a RFC?

I think this will be really useful in two years or so. Thanks for the video! 🙏🏻

What color theme are you using in Cursor?

So the core concept is those "old" tool callings of moderns LLMs, but in a weird new standard. OpenAI did those ChatGPT plugins on top of that, then after changed to function calling inside Custom GPTs. Langchain and the community built a lot of tools based on that too. Context Protocol doesn't seem to be the proper name.

❤❤Thanks for this short overview. aaPanel supports one-click deployment of dify

Still don't understand what this shit is and how you can use it for something else than Claude

"The hottest new programming language is English." ~Andrej Karpathy

Thanks a lot! Working like a charm

As a Suno user, I feel V4 has better sound quality over V3.5, but V3.5 is still more consistent in speech recognition when creating personas. It's improving though. Great video!

how secure is the data that the AI is reading?

The Audio of this Vid is Fake, the speaker doesn't inhale or exhale while speaking

Excellent! Keep up the great communication! Teaching….

How is it compared with Windsurf?

I’ve watched 3 videos, nobody can basically explain what these MCP things do basically

It's basically a purpose built.vector rag type solution, as opposed to using those for the same purpose with less accuracy. I'll be trying it, should be easy to get going

I think you're bringing up some important points in your video by pointing out possible issues with it going in the wrong direction on those longer running agentic tasks. I think that's currently one of the most important issues with Windsurf and the new Cursor agent as well. Keeping AI within the boundaries you want it and not throw over certain parts of your application. I have played around with that quite a bit, trying to let it write some documentation that is mentioned over and over again to have it take care of certain apsects or always be aware of certain goals of the project. My personal experience so far isn't that great - haven't found a reliable way to not break some features over time here and there by not explicitly telling it not to touch it or be careful not to throw things over. In my opinion one of the most challenging problems right now with AI guided development.

I love that you're pointing out this exists. I think it would be even cooler to focus on your take and what it means. What are the implications in your expert opinion? You don't have to be right. I just want to hear what you think.

My programming language will be German 😜 Thanks for the video.

Cursor is terrible compared to Windsurf - composer is a sad rip-off of the WindSurf feature set and now Cursor trying to play catchup.

I mean... I know its the 1st iteration, but if you have to fix everything after every prompt, you are not gaining time but wasting... Personaly i'll stick to the chat window, and we'll see in the future... Thx for the presentation though 👍

That's also my experience with these tools... They fiz one thing and break another. Asking one little thing at a time helps, as well as being very specific about the bugs. Thank you for the great overview!

I am really interested in this because RAG but I doubt what they are saying I have not implemented this in low level code but this seems like another "standard" API that many companies have proposed and have failed to catch on. It's "1 API to rule them all" but it's Anthropic (18 billion) vs Google, Microsoft, Apple, etc, (∞ billion) where Anthropic is telling them what to do. This has never worked and will never work because of something called a "technology moat". The only company that has countered this is Zapier, but only because they wrote 10,000 integrations with different data providers. Please discuss this with me if you are in the know.

Are going to show us how to use this anytime this year ?

Brilliant

How are you using tool calling in claude web?

That music is way too loud

Sounds like a cross between automation tools like n8n and assistants where the assistants can access a variety of data sources without needing custom API integrations

i need features like in Cursor hmu...

I accidentally listened to several songs on the main page and there is no noise and distortion! The quality is almost like that of Udio, even written in version 3

i put in all of my journal entries dating back 7 years and well over 100'000 words. the results were insane, freaky and cool. Having 2 people talk in depth about my evolution over those 7 years and about key moments in my life

"adds type annotations everywhere": actually adds "any" everywhere, which completely does nothing more, then actual simple script would do XD

Many thanks! Is there anyway that we can limit the time of the podcast? I mean making it 5 minutes instead of 10 minutes.

Any luck deploying?

Can we do songs longer than 4 minutes?

The vocals are much better and the compression much better but if it had the same audio quality as udio then suno would be the killer. Suno is good for certain type of music where there is less beats involved. They should add options like no beats.

PLEASE HELP ME GUYS! I wanna remaster a song that i’ve done with a different profile then my main account (where i have a paid subscription with the V4) but i can’t see the “get the whole song” option that i see with others people song in the app home. I’ve made the song public with the other account, but still i don’t see the option. U need another subscription in the other account? Or there’s a way without it? Please let me know 🙏🏻🙏🏻🙏🏻 thanks a lot

mind blowing 🔥

Hmm... to bad I cant create flows myself, like i would like to use my own flows to include vector data for some tasks (like new documentation) or so...

Cool

2500 credits was more than enough for $10 in my opinion. I had so much fun my first month, and I'm excited to see what V4 can do since 3.5 is already really good.