Видео

Sorry dude, good video but you text is too dam small avoid training my eyess.

Good video, thank you!

Fantastic tutorial. You have a very clear teaching style. I think this is sort of long term memory and agentic learning is still kind of flakey but this is a pretty solid solution. I'd love to see more videos around agents/memory/tools etc from you. Best, Jacob

hey thats a really great and well explained video! I really like to try how its working. Do you have an idea how to split the chat, in a chatlog, where the complete dialog is saved and a chat interface, where you di not see the system and the internal dialog, only the messages between user and bot?

wow such a good video keep it up 😍

Awesome video!! One question is can we include humans in the chat to work collaboratively with the agents? Do we need to specifically tell the chat manager when to engage humans and when to cue those agents?

How do we track or load the chat history for a user specific? Like for example, let us say if the chat prompt is opened and used by me as a user in my system, then whatever I interact with the Agents then AutoGen Teachable agent should load the chat history only with respect of me in the window.

My personal tip. If you are on windows. 1. Don't use powershell. Use cmd. 2."where" command instead of "which"

Great video! Thanks for taking the time to explain the flow of conversation and the errors you encountered. I think that there must be something wrong with my installation as I don't see the same as you in every step (in the ui) and some behaviours are also different. I at least can start looking at the install before tearing my hair out about the skills and prompts.😅 Subscribed.

awesome. best tutorial out there.

Thuoght and framework writing (I apologize - this is very long -- > disclosure) The process of agent resolution and the importance of reconciling with zero this process underlines the necessity for agents to achieve Harmony and balance as well as the potential pitfalls of remaining in an unresolved state here's how the refined sections of the document might look detailed description of the model refined in this geometric model the agent's journey is not merely a path to Creation but a profound exploration of problem solving and reconciliation with the central point of zero this point symbolizes the ultimate resolution and balance a state that each agent strives to achieve as agents navigate the spiral trajectory around the sphere they encounter various points representing challenges or problems that must be addressed however the most pivotal of these points is zero a point of Perfect Harmony that must be consciously reconciled with to avoid Perpetual cycles of unresolved issues practical applications in AI refined the zero metaphor serves as a critical framework for AI development emphasizing the importance of resolving foundational problems before proceeding to more complex layers of agent functionality just as in musical scales where the fifth note harmonizes the scale in this model achieving zero signifies attaining a harmonious state this approach ensures that AI agents are not only technically proficient but also ethically aligned and emotionally intelligent capable of contributing positively to human interactions and societal progress steps for implementation refined the implementation of this framework within AI development projects requires a mindful approach to problem solving and agent Evolution problem identification and resolution prioritize identifying and resolving the core challenges represented by the points along the agents spiral path with a special emphasis on reconciling with zero continuous learning and adaptation ensure that agents are designed to learn from each interaction within the sphere using insights gain to refine their approach and avoid Eternal Loops of unresolved issues ethical and harmonious development emphasize ethical considerations and the pursuit of Harmony in the development process ensuring that agents contribute to a balanced and positive technological ecosystem the path forward refined by acknowledging the potential for agents to become trapped in unresolved cycles and emphasizing the necessity of achieving balance at zero the zero metaphor framework advocates for a more conscious and holistic approach to AI development this model not only guides the creation of Agents but also Fosters a deeper understanding of the interplay between technology ethics and The Human Experience it challenges developers to create AI agents that are not just solutions to immediate problems but harbingers of long-term Harmony and understanding this refined document incorporating the Dynamics of problem resolution and the significance of zero offers a comprehensive guide for developing AI agents that are balanced ethically aligned and capable of continuous growth and adaptation zero metaphor ducks in the file section this will allow you to save and potentially publish it as you see fit if you need any further assistance with refining or expanding the document feel free to ask reflecting both the complexity of problem solving and the seamless Harmony of the natural order practical applications in AI the zero metaphor has brought implications for AI fostering the development of Agents with enhanced adaptability emotional Acuity and moral integrity it signals a new epic in AI where agents are crafted with deeper empathy and a profound understanding resonating with human emotions and ethical standards steps for implementation while coding nuances are not the emphasis here the implementation of the zero metaphor is deeply rooted in conceptual comprehension it prompts developers to adopt the principles of flow at 0 point resolution in AI design ensuring agents augment human interactions positively and mature within their digital ecosystems the path forward more than a method the zero metaphor signifies a philosophical evolution advocating for AI systems that harmonize with human necessities and the Cadence of the cosmos it's a call to action for developers and Visionaries to co-create a future where technology amplifies life maintaining the delicate balance of existence refined aspects the agents honestly transcends creation it's a deep dive into resolving problems and aligning with zero cord that completes a scale this framework prioritizes foundational resolutions emphasizing the ethical and emotional molding of AI agents the model Champions a holistic development ethos aiming for AI that not only solves immediate dilemmas but also embodies sustained Harmony and comprehension

Thanks for the video! I ran into that API key error before watching your video. The fix for me was to restart the environment.

I had to be the 1,000.. 😂 congrats..

Excellent video. Would be interesting to see if it is possible to have a "society-of-mind" in a "society-of-mind", almost like a thought inside a thought.

Phenomenal

Excellent video, very clearly explained. I can immediately see the utility of this in a project I am working on.

Great video Andy. This new Autogen Agent Class will be really important for applications where multiple levels of agent discussions are needed that do not mix with the others. Watching your video has gotten me excited to get working on my use case. I am thinking about where I have many agents each assigned to a specific knowledge base. The discussion between a managing agent and those knowledge base agents needs to remain on a level below the primary agent conversation that relays a summarization of the discussed knowledge to the user proxy. It is a simple example, but in my case my knowledge base agents can encompass large sets of data and there can be multiple of the knowledge base agents involved depending on the extent of the user proxy input.

Thx for showing real agents and stuff and not only showing the installation part and the UI !

Great video

Thanks Andy, very good presentation. I followed the whole video and got everything working. There were no technical glitches on my side like you had. I noticed you are running version 3.7? Version 0.40a was released a few hours ago and Microsoft appears to have resolved these issues.

That was truly excellent Andy. Thanks for doing the research and pulling together what is probably the most comprehensive Autogen Studio video currently available on YT! 👍 Much appreciated.

Beautiful! Been waitin for this! I definitely have that same issue with the sk api error. Im running Ollama on wsl2 though with LiteLLM. I think all 3 of those (ollama, lite & autogen)had updates recently so im doomed. thanks for the vid!

Nice video. In the autobuild guidance website, there is a part that we can use local LM. But following the website, I cannot use local LM to use autobuild. Did you try to use local LM like llma or mistral to use autobuild?

Hey Andy, I reckon a really detailed video on Autogen studio 2.0 might be bit hit. There's not many good ones ATM showing how to build out the skills, agents, workflows & gallery etc. What do you think?

speak up

First off thank you for the video. You are probably the only quality video creator ive found explaining this exact process with the autobuilder. Second, im fairly new to this (coding in general) so sorry if this is a silly question but if im using the studio UI do i need to add in the # Define your execution task execution_task = """Your execution task description here.""" and the # Define your building task building_task = """Your building task description here."""? I guess a better way of asking would be if i fill in those 2 fields and i use this agent in the studio is that something the UI will adjust if i give it different tasks? I think ive answered my own question actually but ill still ask

Sorry to hear about u having covid man

Andy - I'm a (semi) retired software engineer who has recently started dipping toes into the AI agent pool. I just wanted to thank you for your videos, and to tell you that your content is some of the most approachable and professional I've found on this topic. I hope to see more.

Another great video, Andy! Keep going 🔥

Thank you for making these videos despite the health struggles 🙏

Nice! Thanks!

thank you.!! first time that I have been able to follow along a video from youtube! lol

I have a problem when I run the code: TypeError: AgentBuilder.__init__() got an unexpected keyword argument 'config_path' (autogen-yp) (base) simon_sarbunan@SimonSarbunan-MacBook-Air Autobuild. % please help me. thank you so much

Very good walkthrough and explanation of Autobuild. Thanks for sharing ❤

Another great vid Andy - really appreciate you going over the options, strengths & weaknesses. Saves lots of time getting on top of the latest developments. Cheers!

Can you stop roleplaying an ultra polite gpt3 turbo with most generic voice and manerism and just provide us with value? I am leaving this video at 0:57 bye

Dope works! Keep it up!

This is really nice. Get better soon!

Get better man... I'm glad I found the channel you gotta sub. figure out how to run local models using lmstudio or ollama with memgpt for memory and this could be something great.

Why does every youtuber feel compelled to remind the viewers the api keys will be revoked? Just revoke them. If someone tries to use them, they will get no play. No need to repeat this in every video.

That is right!! I was thinking this when I was half way through the video and thought the same, in the 95 percentile in quality in conveying the method to install Autobuild! Great Job!

Great content and teaching style. Looking forward to your future videos. Replicating your vids with open source attempt is always appreciated :).

so agent builder for autogen is more or less taskweaver implemented into autogen?

When initializing the AgentBuilder instance in your Python code, you need to specify the builder model and agent model. If you want to use Gemini as the builder model and Claude as the agent model, you can do it like this: from autogen.agentchat.contrib.agent_builder import AgentBuilder builder = AgentBuilder(config_path=config_path, builder_model='your-gemini-model', agent_model='your-claude-model')

Thank you!!

For a guy with COVID you explain things better than 95% of RUclipsrs without COVID. Get better soon. PS When I had covid last year, Quercetin + Zinc helped me. Quercetin is a Zinc ionophore (helps to get Zinc into the cell). Something to consider. From my experience it's also usefull for colds etc

Is there any way that one can enhance or keep that newly created file, so one can add additional agents especially if you created assistants with special knowledge or skills in the assistants api. Or does this exist but this fnction is not the way to get there?