3:48 “The primary skills required for an Al agent developer role can significantly vary from project to project. This topic deserves its own separate video by itself.” Yes it does 👍
Found this channel very late I guess. As a commerce student I'm eager to hop into this journey. Can you pls breakdown on how to get to this role with the specific tools, packages and frameworks to learn ? Thanks a ton
It also means that technical skills, while essential, will need to be paired with a vast life experience if one wants to become a top performer as an AI Agent Developer.
I developed software that has increased the data maintenance process in our shop by around 5000%. Almost everything except quality control is automated. Whereas in the past we needed 2 months to manually set up 1000 items in the shop, for example, we now need maybe 3 days and there is still potential to speed it up even more. I implemented all of this in around 7 months with the help of gpt4 and without any great programming knowledge. There were a few moments when I thought I couldn't do it because the project was already very complex, but I fought through it. I definitely wouldn't have managed it without AI. However, I must also say that I have a bit of experience in development and therefore know which questions I need to ask and have an idea of the goal.
@@PedroHenrique-cu1ur with a lot of c&p, try and error, frustration and passion, I would say. 😂 The road to the goal is paved with failures. There were a few moments where I thought I wouldn't be able to do it because I couldn't solve a problem, but so far I've always managed to make a breakthrough, even if it sometimes took weeks.
@@TobiasKrallmanni’m doing an ai agent project as we speak but i have zero experience with coding and development 😂😂 but im pushing through. I need my own assistant.😂😂
SmythOS sounds like a game-changer for AI automation! The idea of orchestrating teams of AI agents to tackle complex workflows is amazing. The no-code approach really opens up advanced AI capabilities to everyone. Exciting times ahead!
I have been working on training a custom agent for a few months now. I would love to hear anything you have to offer about the skills and techniques involved with training an AI agent. Keep up the great work!
love your videos. I think you guys need an UI. That'll make a world of difference and put your product in the hands of a lot more people who are not just developers.
13:12 i think my question somewhat made it into the video 😜 love your content! came for the diligence, stayed for the passion. i also tried a lot of ai agent frameworks (including crewai and autogen), and also came to the conclusion that i need to build my own; but i also prefer your framework over the others, keep it up!
@@amrinderrandhawa6799i clearly prefer crewai for the very customizable workflows and the ability to not only assign one, but two separate llm's to a single agent if necessary. for me, autogen was more of a "quick and dirty" solution to simpler tasks, but if you need a series of tasks relying on each others outcome and each task requires about 4 agents to be solved properly, you quickly end up with 20+ agents divides into 5+ crews and 5+ tasks to solve, where the quite straightforward "question - response" style of autogen just does not really do it anymore and usually gets pretty messy, as well.
@@amrinderrandhawa6799 clearly crewai; i tried autogen first (mostly because at the time i tried autogen, crewai has not been released yet), but working with autogen feels a lot like "chaos management", whereas crewai is way more focussed, more streamlined and more intuitive imo... both work pretty smoothly if you get your setup going (for autogen, i used olllama and litellm to host multiple llms in parallel on my local devices), which also works with crewai even without the intermediate step of using litellm and both offer very similar capabilities; some of the differences are: autogen offers way more flexibility for agent creation with the "agent class" parameter, which allows to add (to put it videogame-terminology) some sort of "passive skills" - if the ability would be the "active skills" in that regard. - talking about agent classes, one of the most useful classes (at lest for me) would be the "open ai assistant api agent"-class, which can be used as a drop-in replacement for the openai assistant api - which might come in handy depending on your use-case -crewai also offers something that might also come in handy, depending on your use-case: where autogen "only" offers one parameter per agent to set a base url and thus, assign your desired llm to it, crewai actually offers two, the "llm" parameter and the "function calling llm" parameter, enabling each agent to be assigned to two different base urls, if necessary. crewai offers you to define crews and tasks, not only agents, which for me, is the dealbreaker here. as i only (!) use sequential processing without asynchronous tasks, the ability to not only create the agents, but then assign them to different crews, those different crews to tasks and those tasks to processes makes it very "applicable" to real-world use-cases by being very modular by nature. maybe i was just unlucky or i did it wrong, but when you try to make multiple agent-teams interact with each other with them relying on each others outcome, autogen never provided really decent results.... -as stated in the video, as crewai has multiple natural language prompts in the backend, the outcomes might vary depending on your prompt engineering capabilties and (mostly if you want to use crewai with open source llms and not openai) the natural language processing capabilities of the underlying model, as well. (sorry for the wall of the huge wall of text... 😅)
@@amrinderrandhawa6799i prefer crewai over autogen mostly because it focuses more on crews and tasks, whereas autogen seems more agent-focussed (at least, right now...). maybe i just got unlucky or maybe i was doing it wrong, but bigger the agent swarm grew and the more complex the tasks got, the worse autogen performed.
Is there any possible future scope you will support open source LLMs? Shared state between agents should be possible with any database layer, be it in memory or even persisted. Being coupled to one provider is okay in some situations but can be very daunting for other applications and orgs
Amazing, you are right Agents is the next HR industry and there will be a lot of maintenance and cleaning to do with all the virtual rubbish we all are creating not knowing exactly what we are doing, I am interested in developing emotionally intelligent agents. I have some ideas, but most wont get it... I know it is possible because emotions are a construct, and constructs are based on conditions and conditions, once recognised can be written, and measured. I should know I am a psych.
00:06 - AI agents will be hired by businesses in the future 03:04 - AI Agents possess decision-making capabilities 05:47 - 2.0 AI Agent-Based applications grant autonomy to agents and allow them to adapt accordingly. 08:28 - Splitting tools into multiple agents eradicates confusion and allows for easier scalability 10:53 - Overview of agent communication in AI development 13:39 - Agent state management is crucial for building AI agents efficiently 16:14 - Use Pydantic for data validation in building AI agents. 18:46 - Creating AI agents with hierarchical communication flows 21:11 - Fine-tuning Image Generator tool for using AI agents 23:26 - Building AI agents to streamline Facebook ad creation process. 25:58 - Establishing communication flows between AI agents 28:20 - Apply for jobs with previous experience in the framework
This manual is a great help for newbies! Multi-agent systems are powerful but difficult. Tools like SmythOS are making them more accessible. Developing news protocols between agents affects me the most. Have you had any knowledge building multi-agent systems? What challenges did you face
was skeptical because it seems like a sales pitch, but it really is very useful. going to use it, i've been trying to make something like this from scratch.
Great video, thanks. I want to create a data analysis agent but the problem I have been facing is implementing reasoning capabilities. Essentially I want first an agent to create a plan how it should answer a user query. Then another agent that follows that orchestrate that plan by dividing the tasks to different agents when tasks are independent, and for dependent tasks, it needs to wait for other tasks that finish first and then provide the results to a new agent to finish that task. All in all, I cannot get my head around how I can implement such a system. Any tips will be highly appreciated.
I think you really get it. You mentioned being critical of langchain; of which I've been looking it to. Could you offer a more detailed critique of the framework?
Curious as to what agents cost to have them built for you, as you said, one at a time? Thanks for the content! This stuff has me blown away! So amazing!
I'm loving the agency_swarm framework. Really great work. Can you provide an example of using the framework with front-ends other than Gradio (with streaming)? Thanks in advance!
Wow wt* I was using autogen and other stuff, but couldn’t get it how I wanted it… I was planning to create it all by myself because maybe that’s easier then get it to work… I saw this video, and I watch this video and see every issue I had with it is solved INSANE!! Hihi. Love this thank you so much, for making this❤️❤️❤️
Nice video and amazing framework, thanks for it... Do you know how to force a agent to iterate multiple times over a tasks? Let's say that I want to write a full long article for a blog... I have a agent that creates the structure for the article, let's say 1. introduction 2. what is agency swarm? 3. what is a AI agent? 4. how to set up a team of agents? etc etc how can I tell the agent writer to write only 1 section at time? instead of the entire article? thank you
I could not reproduce. I had a lot of errors. No module named 'readline', module 'collections' has no attribute 'Callable', "The requested model 'gpt-4-turbo-preview' does not exist. etc...
'readline' => i also had this problem, I fixed it with: pip install pyreadline3 ...on/for windows. And: import collections collections.Callable = collections.abc.Callable ...in python code
I want to be able to adjust settings like temperature, token limit, stop sequences and frequency. This can all be done in OpenAI. Can I integrate this through to the chatbot widget? This is essential. Voiceflow has this. :)
Hi how can I incorporate this into a ai call center like vapi? So that it takes care of my entire business for me. I have a construction business and I want it to also relay orders to my staff and answer customers? Is there an updated version
How are you handling proprietary company data that the company does not want to send/share to OpenAI or leave its company boundaries? In my line of work this is critical and why open-source shines as I can keep all data internal to the company.
@@vrsen thanks I will check that out. In the gov space this is a huge dealbreaker. If using Azure OpenAI, I also see why it doesn't make sense to support open-source models. Thanks
Quickly question: How do I utilize OpenAI and still be GDPR compliant? So far I believe that AWS and Llama 3 for example would provide the best security regarding gdpr compliance. Can you guys give me some feedback on how to achieve the highest GDPR compliance? :) I am located in Germany, here it really is a big thing
I appreciate the attempt to transformr the AI agent framework landscape but why do we have to upload our files into Open AI, this is not appropriate for serious companies and enterprises as they will not want to share their documents to open AI, unless I have misunderstood something. E.g are we able to pull data from azure or other data stores? ALso how do you then implement azure open AI's API rather then the open AI assistance?
the point the guy in the video makes is that open source is not free, there’s no api fees but you still have to pay for your own hardware/infrastructure to run something at scale. is that not true?
0:52 It’s not just a cleverly prompted LLM. That’s factually incorrect. It’s an LLM that was further trained to improve its long term planning and reasoning abilities. Which is not an easy task to do.
If by trained you mean actual training, like adjusting the weights, than It’s not trained. They are using publicly available llms like Claude and GPT 4. I will try to beat them in a few videos on SWE.
"The journey of AI, according to SmythOS, is a philosophical exploration that unveils deeper insights into consciousness and human identity with each technological milestone."
Hey guys! I'm immersed in the study of AI agents and I'm curious: would it be viable to build an agent that prospects customers for freelance professionals? I envision a system capable of exploring Instagram in an automated way, identifying potential customers and even starting conversations to schedule sales meetings. Is it possible to develop such AI agents? If so, do you know of any videos on RUclips or any mentors that explain how to create AI agents to automatically prospect customers through Instagram? Is there something like this in development or is this an idea for the future of AI?"
If Langchain was made after the Assistants API, would it be a breakthrough solution? I think that in general we could see something like Agency Swarm, but with a different LLM, although I understand that we can't replace Instructor, right?
I have a bit of a noob question, I have used the genesis feature to spin up my agency, but now I am getting the "OpenAI key not set" with the message " ValueError: OpenAI API key is not set. Please set it using set_openai_key." Which file do I set this in
Is this currently useable with offline LLMs and if not do you intend on making it capable of doing so? I couldn't find documentation anywhere on directing your framework to a local LLM.
pip install worked but agency-swarm genesis gave some issues, according to gpt it seems like there is still an issue with the readline module, specifically related to compatibility issues with Python 3.10. The error message indicates that there is an attribute error in the py3k_compat.py file of the pyreadline package. This issue might be due to the pyreadline package not being fully compatible with Python 3.10 yet. Which version of python is compatible? i've so far tried 3.9, 3.10.10 3.11 and latest with no luck. Would be cool to be able to get past the second step😂
What I really like about agency swarm from a conceptual standpoint is We can elevate an organizational model of how groups interact and use the framework to model how scientist and cognitive researchers think the human mind works Imagine if we can create agents that are essentially a collection of various processes that mimic what we think happens within people. We have conscious awareness, subconscious processes, and autonomous systems. All working underneath. The consciousness of our awareness yet are happening regardless. This sort of structure and thinking of the system can easily be extrapolated and presented as an entirely different use case outside what Arsen originally planned his solution to be.
I own an ecommerce store selling catering equipment. Do you think it would be possible to get ai to do everything or almost everything? If so, guess I must reach out to you!
If you can show us how to integrate this locally (for example using LM studio for the LLM and Stable Diffusion for the image generation for the ad agency), the video will EXPLODE>
The Previous video explained, the challenge so use hi solution to try and hardcode those systems or get him to do it for you, which is part of the service he offers what's your use case let me see if I can help.
Hi , amazing concepts , but can in your agency swarm can you add multiple llm support using langchain. i tried to do it on my fork . but it was getting complicated. Would love a video from you ooon this topic
You can import tools form langchain, see “advanced usage” from documentation. Although, obviously, I do not recommend this. Use tool creator gpt instead
@@vrsen advanced usage from agencyswarm or langchain ?. Btw can you list your reasons for agency swarm not being llm agnostic in a seperate video. going into the nitty gritty details. i k you covered some reasons for that , but I think people would like to know more about this. And maybe some ways to make this work with other llms
I think this video and the previous ones sufficiently explain why the bet was made on the Assistants API, particularly in terms of state management. And within 3-4 months, OpenAI will release a new model, making the agency swarm both cheaper and better🎉
The innovative approach of *Phlanx's caption generator* transforms social media management into a profit-generating activity. By enhancing engagement and expanding reach, businesses can attract new customers and retain existing ones more effectively. This direct impact on revenue makes it a must-have tool for any business looking to thrive in the digital age.
@Vsern thanks for this tutorial, i would like to borrow some of your ideas for my thesis but trying to follow through your video i am facing challenges with this error, 'File "C:\Users\hp\AppData\Local\Programs\Python\Python310\lib\site-packages\openai\_base_client.py", line 988, in _request raise self._make_status_error_from_response(err.response) from None openai.BadRequestError: Error code: 400 - {'error': {'message': "The requested model 'gpt-4-turbo-preview' does not exist.", 'type': 'invalid_request_error', 'param': 'model', 'code': 'model_not_found'}}', kindly assist with some insights on how to go about it. thanks
3:48 “The primary skills required for an Al agent developer role can significantly vary from project to project. This topic deserves its own separate video by itself.” Yes it does 👍
Got it!
+1
+2
Found this channel very late I guess. As a commerce student I'm eager to hop into this journey. Can you pls breakdown on how to get to this role with the specific tools, packages and frameworks to learn ? Thanks a ton
It also means that technical skills, while essential, will need to be paired with a vast life experience if one wants to become a top performer as an AI Agent Developer.
I developed software that has increased the data maintenance process in our shop by around 5000%. Almost everything except quality control is automated. Whereas in the past we needed 2 months to manually set up 1000 items in the shop, for example, we now need maybe 3 days and there is still potential to speed it up even more. I implemented all of this in around 7 months with the help of gpt4 and without any great programming knowledge. There were a few moments when I thought I couldn't do it because the project was already very complex, but I fought through it. I definitely wouldn't have managed it without AI. However, I must also say that I have a bit of experience in development and therefore know which questions I need to ask and have an idea of the goal.
What's Ur email let's talk business
It's great that your company is developing it, how did you do this using GPT-4?
Yep. The insights and value are knowing what questions to ask. Tech does replace humanity, it supplements and compliments it...nice work by you
@@PedroHenrique-cu1ur with a lot of c&p, try and error, frustration and passion, I would say. 😂 The road to the goal is paved with failures. There were a few moments where I thought I wouldn't be able to do it because I couldn't solve a problem, but so far I've always managed to make a breakthrough, even if it sometimes took weeks.
@@TobiasKrallmanni’m doing an ai agent project as we speak but i have zero experience with coding and development 😂😂 but im pushing through. I need my own assistant.😂😂
As a PHP / JS developer by day, I am going to reboot my langchain course at night after watching this. - very inspiring stuff mate!
SmythOS sounds like a game-changer for AI automation! The idea of orchestrating teams of AI agents to tackle complex workflows is amazing. The no-code approach really opens up advanced AI capabilities to everyone. Exciting times ahead!
I have been working on training a custom agent for a few months now. I would love to hear anything you have to offer about the skills and techniques involved with training an AI agent. Keep up the great work!
@5:10 this is gold. I tried explaining this to a tech lead and wish I’d had this video so they could hear it not just from me
Bro this is unbelievable this is the exact video I was looking for when tapped on the youtube app... thanks as this is greatly appreciated...🎉
love your videos. I think you guys need an UI. That'll make a world of difference and put your product in the hands of a lot more people who are not just developers.
Yup, working on it
Hey arun, have you tried building any ai agent?
Love your content and all your work! Thanks for this guide! Making a video on AI Agent Developers would be awesome.
The best overview I have seen yet. Very helpful. Thank you!
are you the creator? this info is trash, he said absolutely nothing
@@zacki5663 I am not the creator. Not sure why you would think that.
Great video, thank you, got my agency genesis up and running🎉🎉
13:12 i think my question somewhat made it into the video 😜
love your content! came for the diligence, stayed for the passion.
i also tried a lot of ai agent frameworks (including crewai and autogen), and also came to the conclusion that i need to build my own; but i also prefer your framework over the others, keep it up!
which one did you like more between crewai and autogen ?
@@amrinderrandhawa6799i clearly prefer crewai for the very customizable workflows and the ability to not only assign one, but two separate llm's to a single agent if necessary.
for me, autogen was more of a "quick and dirty" solution to simpler tasks, but if you need a series of tasks relying on each others outcome and each task requires about 4 agents to be solved properly, you quickly end up with 20+ agents divides into 5+ crews and 5+ tasks to solve, where the quite straightforward "question - response" style of autogen just does not really do it anymore and usually gets pretty messy, as well.
@@amrinderrandhawa6799 clearly crewai; i tried autogen first (mostly because at the time i tried autogen, crewai has not been released yet), but working with autogen feels a lot like "chaos management", whereas crewai is way more focussed, more streamlined and more intuitive imo...
both work pretty smoothly if you get your setup going (for autogen, i used olllama and litellm to host multiple llms in parallel on my local devices), which also works with crewai even without the intermediate step of using litellm and both offer very similar capabilities; some of the differences are:
autogen offers way more flexibility for agent creation with the "agent class" parameter, which allows to add (to put it videogame-terminology) some sort of "passive skills" - if the ability would be the "active skills" in that regard.
- talking about agent classes, one of the most useful classes (at lest for me) would be the "open ai assistant api agent"-class, which can be used as a drop-in replacement for the openai assistant api - which might come in handy depending on your use-case
-crewai also offers something that might also come in handy, depending on your use-case: where autogen "only" offers one parameter per agent to set a base url and thus, assign your desired llm to it, crewai actually offers two, the "llm" parameter and the "function calling llm" parameter, enabling each agent to be assigned to two different base urls, if necessary.
crewai offers you to define crews and tasks, not only agents, which for me, is the dealbreaker here. as i only (!) use sequential processing without asynchronous tasks, the ability to not only create the agents, but then assign them to different crews, those different crews to tasks and those tasks to processes makes it very "applicable" to real-world use-cases by being very modular by nature. maybe i was just unlucky or i did it wrong, but when you try to make multiple agent-teams interact with each other with them relying on each others outcome, autogen never provided really decent results....
-as stated in the video, as crewai has multiple natural language prompts in the backend, the outcomes might vary depending on your prompt engineering capabilties and (mostly if you want to use crewai with open source llms and not openai) the natural language processing capabilities of the underlying model, as well.
(sorry for the wall of the huge wall of text... 😅)
@@amrinderrandhawa6799i prefer crewai over autogen mostly because it focuses more on crews and tasks, whereas autogen seems more agent-focussed (at least, right now...). maybe i just got unlucky or maybe i was doing it wrong, but bigger the agent swarm grew and the more complex the tasks got, the worse autogen performed.
hey i would like to know ,can i create this crew agent & all completely free ??most of them need some sort of Open ai API or some paid services?
Is there any possible future scope you will support open source LLMs? Shared state between agents should be possible with any database layer, be it in memory or even persisted. Being coupled to one provider is okay in some situations but can be very daunting for other applications and orgs
Agreed
Amazing, you are right Agents is the next HR industry and there will be a lot of maintenance and cleaning to do with all the virtual rubbish we all are creating not knowing exactly what we are doing, I am interested in developing emotionally intelligent agents. I have some ideas, but most wont get it... I know it is possible because emotions are a construct, and constructs are based on conditions and conditions, once recognised can be written, and measured. I should know I am a psych.
Great video, love the detailed explanation. So smart.
i just fell in love with this channel. great vids! you just got a new sub! 😘
00:06 - AI agents will be hired by businesses in the future
03:04 - AI Agents possess decision-making capabilities
05:47 - 2.0 AI Agent-Based applications grant autonomy to agents and allow them to adapt accordingly.
08:28 - Splitting tools into multiple agents eradicates confusion and allows for easier scalability
10:53 - Overview of agent communication in AI development
13:39 - Agent state management is crucial for building AI agents efficiently
16:14 - Use Pydantic for data validation in building AI agents.
18:46 - Creating AI agents with hierarchical communication flows
21:11 - Fine-tuning Image Generator tool for using AI agents
23:26 - Building AI agents to streamline Facebook ad creation process.
25:58 - Establishing communication flows between AI agents
28:20 - Apply for jobs with previous experience in the framework
This manual is a great help for newbies! Multi-agent systems are powerful but difficult. Tools like SmythOS are making them more accessible. Developing news protocols between agents affects me the most. Have you had any knowledge building multi-agent systems? What challenges did you face
Great video, and yes, please make a video on AI Agent Developers. Thank you!
was skeptical because it seems like a sales pitch, but it really is very useful. going to use it, i've been trying to make something like this from scratch.
I've never heard an Eastern European say "nitty-gritty", but now I'm hooked. Please say it more.
Great video, thanks. I want to create a data analysis agent but the problem I have been facing is implementing reasoning capabilities. Essentially I want first an agent to create a plan how it should answer a user query. Then another agent that follows that orchestrate that plan by dividing the tasks to different agents when tasks are independent, and for dependent tasks, it needs to wait for other tasks that finish first and then provide the results to a new agent to finish that task. All in all, I cannot get my head around how I can implement such a system. Any tips will be highly appreciated.
I think you really get it. You mentioned being critical of langchain; of which I've been looking it to. Could you offer a more detailed critique of the framework?
Hard to customize because there are too many unnecessary abstractions for features that can be implemented by yourself in 5 minutes
The knowledge is crazy you earned yourself a sub
Go go boss. Always growing!
Curious as to what agents cost to have them built for you, as you said, one at a time?
Thanks for the content! This stuff has me blown away! So amazing!
I'm loving the agency_swarm framework. Really great work. Can you provide an example of using the framework with front-ends other than Gradio (with streaming)? Thanks in advance!
Awesome video , thank you for sharing your knowledge.
Wow wt* I was using autogen and other stuff, but couldn’t get it how I wanted it… I was planning to create it all by myself because maybe that’s easier then get it to work…
I saw this video, and I watch this video and see every issue I had with it is solved INSANE!! Hihi.
Love this thank you so much, for making this❤️❤️❤️
Nice video and amazing framework, thanks for it...
Do you know how to force a agent to iterate multiple times over a tasks?
Let's say that I want to write a full long article for a blog...
I have a agent that creates the structure for the article, let's say
1. introduction
2. what is agency swarm?
3. what is a AI agent?
4. how to set up a team of agents?
etc etc
how can I tell the agent writer to write only 1 section at time? instead of the entire article?
thank you
I could not reproduce. I had a lot of errors. No module named 'readline', module 'collections' has no attribute 'Callable', "The requested model 'gpt-4-turbo-preview' does not exist. etc...
ChatGPT
@@vrsen I solved some problems using it, but I could not find how to specify the model to use.
'readline' => i also had this problem, I fixed it with: pip install pyreadline3 ...on/for windows.
And:
import collections
collections.Callable = collections.abc.Callable
...in python code
I want to be able to adjust settings like temperature, token limit, stop sequences and frequency. This can all be done in OpenAI. Can I integrate this through to the chatbot widget? This is essential. Voiceflow has this. :)
Hi how can I incorporate this into a ai call center like vapi? So that it takes care of my entire business for me.
I have a construction business and I want it to also relay orders to my staff and answer customers?
Is there an updated version
How are you handling proprietary company data that the company does not want to send/share to OpenAI or leave its company boundaries? In my line of work this is critical and why open-source shines as I can keep all data internal to the company.
If you deploy on Azure, all you data stays on your own private instance. vrsen.github.io/agency-swarm/advanced-usage/azure-openai/
@@vrsen thanks I will check that out. In the gov space this is a huge dealbreaker. If using Azure OpenAI, I also see why it doesn't make sense to support open-source models. Thanks
@@playboy71322why his framework doesn't support open source models and what makes it different from crewai?
Quickly question: How do I utilize OpenAI and still be GDPR compliant?
So far I believe that AWS and Llama 3 for example would provide the best security regarding gdpr compliance.
Can you guys give me some feedback on how to achieve the highest GDPR compliance? :)
I am located in Germany, here it really is a big thing
Hey, I am also located in germany and would love to connect regarding this problem, how may I reach out to you?
Very Nice structure and well delivered.
A sub was created.
Thank you
You are insane this is awesome work!
I don't understand how the code works but I feel like this is important 😅
Hi! Could you please share the type of the keyboard you're using ? It looks pretty solid and comfortable. Am I wrong? Cheers!
This is really cool, thanks for doing this.
Great content! Does it give support to Vertex AI and Gemini?
How would you link this up with a nextjs interface instead of gradio?
ok that was insightful, but what do we need to learn before all this? From a person that did Python course for 1-2 months. Thanks!
Bro I have no clue how to code, so I'm clueless and lost watching these AI agent tutorials. Is zapier or bubble no code the best route for me?
Just learn all you can - start with Python and wait because coding won't be needed anymore.
I appreciate the attempt to transformr the AI agent framework landscape but why do we have to upload our files into Open AI, this is not appropriate for serious companies and enterprises as they will not want to share their documents to open AI, unless I have misunderstood something. E.g are we able to pull data from azure or other data stores?
ALso how do you then implement azure open AI's API rather then the open AI assistance?
Can we use local LLMs with this framework?
Watch the video.
no
imagine having morning standup with your agents ☠
This is actually going to be the case after OpenAI releases memory feature in Assistants API.
@@vrsen its gonna get crazy 😎
Nah, i'll stick to open source. Good luck though!
can you elaborate on why? not challenging, just a noob wanting to hear another perspective
@@outsidesoundsOpenAI is not open 😭🤣, why you ask that? Everyone know if you want to use OpenAI api you need to spend money
the point the guy in the video makes is that open source is not free, there’s no api fees but you still have to pay for your own hardware/infrastructure to run something at scale. is that not true?
@@outsidesounds open source mean the source code or source data is free. Meanwhile OpenAI never reveal which data they are using.
Free Open Source Software (FOSS) IS FREE. OpenAI is not FOSS
This dude is genuis !
How do we account the token counts, in production it is must!!!!
0:52 It’s not just a cleverly prompted LLM. That’s factually incorrect. It’s an LLM that was further trained to improve its long term planning and reasoning abilities. Which is not an easy task to do.
If by trained you mean actual training, like adjusting the weights, than It’s not trained. They are using publicly available llms like Claude and GPT 4.
I will try to beat them in a few videos on SWE.
@@vrsen Nope. They are not using out-of-the-box LLMs.
The Agent is the New Function.
I literally see a time where functions are replaced by agents.
3:56 definitely interested in seeing thay video!
"The journey of AI, according to SmythOS, is a philosophical exploration that unveils deeper insights into consciousness and human identity with each technological milestone."
Hey guys! I'm immersed in the study of AI agents and I'm curious: would it be viable to build an agent that prospects customers for freelance professionals? I envision a system capable of exploring Instagram in an automated way, identifying potential customers and even starting conversations to schedule sales meetings. Is it possible to develop such AI agents? If so, do you know of any videos on RUclips or any mentors that explain how to create AI agents to automatically prospect customers through Instagram? Is there something like this in development or is this an idea for the future of AI?"
Can you build this thing for a person for their individual business needs?
If Langchain was made after the Assistants API, would it be a breakthrough solution? I think that in general we could see something like Agency Swarm, but with a different LLM, although I understand that we can't replace Instructor, right?
I have a bit of a noob question, I have used the genesis feature to spin up my agency, but now I am getting the "OpenAI key not set" with the message " ValueError: OpenAI API key is not set. Please set it using set_openai_key." Which file do I set this in
This is amazing !
are u a dev or are u more of business owner idea guys that understands ai functionality and have devs team ?? am just curious I love your content ty!
the installation of Agency Swarm assumes the usage of GPT 4 instead of 3.5. How to choose GPT 3.5?
Great topic, thanks 👍
Talks about agent swarms and scalability... Lets have more..
Right after ...
9:15 but we will code them one at a time 😅😂
Great work, please keep on
Great video, thank You.
Thank you!
Is this currently useable with offline LLMs and if not do you intend on making it capable of doing so? I couldn't find documentation anywhere on directing your framework to a local LLM.
Local LLM is hard though. No direct documentation, but you can learn how to use model in local.
the developer clearly states, no open source models, only OpenAI ChatGPT models. OpenAi does not make any production open source free models.
@@drnotebook he came out with a video shortly after this one saying that he changed his mind and in fact made it so you CAN use offline LLMs
Please make a video of the AI agent developer levels on project basis
You got it!
You seem to be against usage of langchain.. is it me surmising you wrong or am I correct?
Is that Visual studio code your using there @vrsen ?
pip install worked but agency-swarm genesis gave some issues,
according to gpt
it seems like there is still an issue with the readline module, specifically related to compatibility issues with Python 3.10. The error message indicates that there is an attribute error in the py3k_compat.py file of the pyreadline package.
This issue might be due to the pyreadline package not being fully compatible with Python 3.10 yet.
Which version of python is compatible? i've so far tried 3.9, 3.10.10 3.11 and latest with no luck. Would be cool to be able to get past the second step😂
I'm interested in doing and learning like you ... but for now there is no budget. just started business ... can help
Very interested in AI agent developer course
Awesome progress
What I really like about agency swarm from a conceptual standpoint is We can elevate an organizational model of how groups interact and use the framework to model how scientist and cognitive researchers think the human mind works
Imagine if we can create agents that are essentially a collection of various processes that mimic what we think happens within people. We have conscious awareness, subconscious processes, and autonomous systems. All working underneath. The consciousness of our awareness yet are happening regardless. This sort of structure and thinking of the system can easily be extrapolated and presented as an entirely different use case outside what Arsen originally planned his solution to be.
Can I fine-tune this models on my data or use fine tuned models?
Yes you can use fine tuned models on your openai account by changing model parameter in your Agent classes
@@vrsen Thank you!
I own an ecommerce store selling catering equipment. Do you think it would be possible to get ai to do everything or almost everything? If so, guess I must reach out to you!
Crazy how he talks about gpt 3.5 turbo and it feels… old. Like, it was 3 months ago lol
Yeah we are way past that ,. 5 will be here soon
Interested in the skills needed for an AI developer agent
I was working on crew ai to mimic work of SDET engineer, I think I should try swarm now
👍
cool thanks for the new video!
If you can show us how to integrate this locally (for example using LM studio for the LLM and Stable Diffusion for the image generation for the ad agency), the video will EXPLODE>
Watch the video.
The Previous video explained, the challenge so use hi solution to try and hardcode those systems or get him to do it for you, which is part of the service he offers what's your use case let me see if I can help.
Great!!! Thanks
Which program is he using?
visual studio code
Did this video age well or has these New models wiped out all these.
underated
Yes, AI agent development Role video please...
Can I subscribe twice to this channel 🤔🤗🤗
hey ,do i need to purchase some some services while creating this. agents ??
If you want to use gpt-4 you need to buy openai credits
Hi , amazing concepts , but can in your agency swarm can you add multiple llm support using langchain. i tried to do it on my fork . but it was getting complicated. Would love a video from you ooon this topic
You can import tools form langchain, see “advanced usage” from documentation. Although, obviously, I do not recommend this. Use tool creator gpt instead
@@vrsen advanced usage from agencyswarm or langchain ?. Btw can you list your reasons for agency swarm not being llm agnostic in a seperate video. going into the nitty gritty details. i k you covered some reasons for that , but I think people would like to know more about this. And maybe some ways to make this work with other llms
I think this video and the previous ones sufficiently explain why the bet was made on the Assistants API, particularly in terms of state management. And within 3-4 months, OpenAI will release a new model, making the agency swarm both cheaper and better🎉
Is this real person ? of a AI generated character / content. ?
Great! Is It possible to use local LLM?
the developer clearly states, no open source models, only OpenAI ChatGPT models. OpenAi does not make any production open source free models.
I am kinda new to this, but i want to learn... can you teach new comers?
Please i also want a separate video for the skill necessary for AI Agent developer, please
FYI McDonalds does not hire cooks.
What do u mean?😂
The innovative approach of *Phlanx's caption generator* transforms social media management into a profit-generating activity. By enhancing engagement and expanding reach, businesses can attract new customers and retain existing ones more effectively. This direct impact on revenue makes it a must-have tool for any business looking to thrive in the digital age.
I couldn't even get past step one LOL , I do pip install --agency swarm and nothing happens.
pip install agency-swarm 🤣
Круто! Очень круто!
@Vsern thanks for this tutorial, i would like to borrow some of your ideas for my thesis but trying to follow through your video i am facing challenges with this error, 'File "C:\Users\hp\AppData\Local\Programs\Python\Python310\lib\site-packages\openai\_base_client.py", line 988, in _request
raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'message': "The requested model 'gpt-4-turbo-preview' does not exist.", 'type': 'invalid_request_error', 'param': 'model', 'code': 'model_not_found'}}', kindly assist with some insights on how to go about it. thanks
the request model didn't exist
Can you make a video about how to make an AI agent that can build AI agents?
I’m interested in Ai developer video
Can i steal you algorithm and implemented it to open source model?