Before enabling Claude Dev, I need to enter the Anthropic API Key. Do I have to use the paid upgraded API Key? Or can the free level API Key still run normally?
Thanks for the video but i fail to understand the purpose of building an Ai Chat Agent. These things already exist in the market. What's stopping me or any other client from just buying a top tier full fledged tier/plan from existing players such as: Dialogflow , Zendesk, Intercom, Tidio?
why even buy when you can use your own servers to host the ai chat agent? i dont understand why people like to spend money with no reason. You already have a tone of ready made examples using databases rag etc to make your own even with streamlit.....i dont really understand why ?
Well done for the excellent works. I would like to kindly suggest if is possible show us how to automate a posting on instagram. I mean crafting a required content via Autogen and then automatically posting it.
I coded up a webapp that utilises the open api call for text to speech, speech to text, and vision. Did most of the work with gpt4o and used claude dev to polish it all off. Its a great piece of tech but expensive
The process of creating AI chat bots is getting easier. For individuals seeking a more customized approach, SmythOS provides AI agents that can be precisely adjusted to meet your client's unique needs, guaranteeing a flawless match each and every time.
there are a lot of things you talk about that I don't understand I would like to talk to you first about transcript videos and thanks to his videos I would like to transform gpt into 1 expert teacher in the subject covered in this tutorial and he explains to me all the details discussed so he can make me a script of your video but with 10 times more words and what would be great is if there were also videos and images There are several ways to retrieve articles and therefore have summaries made by chatgpt Do you do this with videos too? There are GPTs that allow you to retrieve the scripts of a video For example, would it be a good idea to take one of these GPTs and put it in a make scenario? Also is there a difference between a custom and a gpt assistant? And an extension? There is a youtube sumarize extension to summarize RU-vid videos, do you think it's different from a custom GPT? It is possible to make your own extensions from a prompt, have you ever done that? Is an extension like an application? What I would like is to take the RU-vid sumarize extension or a GPT to retrieve the transcript of one or more videos and to tell it in addition to giving me a summary to behave like a teacher and to dissect everything and explain everything to me with lots of examples so that I can clearly understand the video I am watching also I don't understand why the chrome extension I'm talking about or a GPT can retrieve the script of a video whereas when I'm in the chat and I give it the URL of a video and I ask it to retrieve the transcript of one or more videos he tells me he can't? is it a question of API, API keys? I don't think so since when I use the extension it doesn't ask me for any key also is there a difference between a GPT and a summarize extension? By creating my summary GPT, would it allow me to do additional things that the current GPTs do not allow me to do? also rather I was talking about the intelligence of gpt that is to say to transform him into an expert and teacher of the subject covered in this or that video, so tomorrow do a long tutorial explaining all the details that you do not explain to me by example in your videos but so I told you that he is not capable of doing it in the chat in any case I understood that I could make it much more intelligent with a custom GPT or more precisely by creating a small database or perhaps even an action with API I watched your video on RAG but I didn't understand anything, I don't have the necessary level Would a rag or a custom GPT allow me to achieve my goals? anyway thank you very much for your videos
I saw that when it was presenting you with the limits and the rate limits, was that due to essentially there being a tear in pricing options? So for example, if you had agreed to pay more or chosen a higher plan, then would there have been no more rate limits? Or are there going to be rate limits regardless of whatever anthropic plan you would be on? Thanks for the video very informational!
Yeah, watching this video and seeing him run out the rate limit so quickly and already at $1 after just a few minutes of tinkering. Seems like it might be better building out everything first using a pro version of the AI and then after having everything initiallly set up, then going to this API solution for debugging and refactoring. For doing the entire app from end to end, though, no way.
Just get yourself and account on AWS, it significantly increases your limits. Agree though, that I'd use the chat version to get the basics in place before pumping up the tokens.
I like your style of editing I use ai, but don't feel like i want so much content only about Ai. Do hou have other YT projects (channels)? I would like more videos with your style of presenting information on different topics besides Ai Thank you ♥️
The reason why you reached $1 so quickly was because you didn’t split up the tasks. The more you ask it, the more the context window becomes larger and chugs on the tokens per task (TPT)
the amount of compute and electricity used by AI building the snake game only this last year is insane 😅, it's a benchmark to test AI but who doesn't know how to code such simple games after being programming for a while
@@ai-for-devs nice lol. qq pls, i tried creating anthropic console account for api key, but this is only for organisation so far, it's not valid for individual use. Did you create an account for an organisation? Thanks
That’s a great question. I think it’s necessary to manage limited capacities and ensure the system runs smoothly for everyone. It’s definitely frustrating, though, especially when you’re paying for the service.
I've used it to build custom scripts for certain cybersecurity tools and yes,it's extremely capable. Over the next 5 years,this type of stuff will redefine how we do work
I’m pretty sure they’ll adjust this in the future. The current token limit can be a real creativity blocker, but as demand grows, they’ll likely adapt to better meet developers’ needs. Fingers crossed!
For client work, the investment can pay off. If these tools can double productivity, it’s a win-win. I’m looking forward to seeing how Aider and open-source models stack up in an upcoming video. It should be an interesting comparison!
It shouldn't. I also have an M.Sc. in Computer Science, and even after a decade, it's still helpful that they taught me algorithms and data structures to understand and evaluate results that are produced by agents.
It wasn't described in the video but SaaS means Software as a Service. The second project of the CV analysis could be deployed as a SaaS project, meaning that people pay to use the software, usually on a subscription basis. Technically the SaaS components of registration and subscription handling wasn't developed but the CV analysis software "could" be developed into a SaaS project.
could you add a 2nd video that includes a llm chat interface that allows user to manipulate the transcript to change mis-spellings like people, places, addresses? and then maybe a pdf export feature? thank you, subscribed!