🔥Join the AI Engineer Bootcamp: Hey there! The second edition of the AI Engineering Cohort is starting soon 🚀 - Learn with step-by-step lessons and exercises - Join a community of like-minded and amazing people - I'll be there to personally answer all your questions 🤓 - The spots are limited since I'll be directly interacting with you You can join the waitlist now 👉 course.alejandro-ao.com/ Cheers!
Beautiful of pure Python - it gives you really deep understanding unlike frameworks! Thank you, man! You have great talent of explaining everything in simple way!
Excellent video! You mentioned the JSON format towards the end. Actually, you can ask for a JSON directly in the prompt. I’ve tried it, and it works really well!” Example session: Question: What is the mass of Earth times 2? Thought: I need to find the mass of Earth Action: {"action_name": "get_planet_mass", "param": { "planet": "Earth" }} PAUSE
@@gazzalifahim The only difference in the prompt is already mentioned in my comment. Actually, the credit goes to Hasan Aboul Hasan; this is his example: ruclips.net/video/cDm5vPXVln8/видео.html&ab_channel=HasanAboulHasan
indeed! you can do this with function-calling LLMs. i am adding this to the next video in the series. was gonna upload it before but had been super busy working on a cohort coruse that i'm doing with some subscribers!
@@alejandro_ao I am trying to sign up but I am trying to learn more of the curriculum that will be covered in it. Could you provide more details? I couldn't find it on the website
Hey! I am going to start teaching and I will be taking a few students. In short, it is 12 weeks of weekly lectures, exercises and live QAs. We will cover GenAI, LangChain, CrewAI, LangGraph and deployment. I made a quick landing page to explain what the program is: course.alejandro-ao.com/ I would appreciate your feedback about the landing page, as I am not great at marketing 😅
It would be great to have a video about utilizing ReAct in LlamaIndex after this one. A very advanced one with query rewriting (if possible) and mainly the ReAct in the LlamaIndex. To me, the document load and parse are easy to pick up but ReAct is confusing.
@@alejandro_ao I know about LlamaIndex from your videos. I'm very impressed with LlamaParse, especially when they split the document by pages for PDF and presentation files. It's exactly how we consume documents in real life; we read page by page. I mean, it's very natural.
hey there. for production code, i would probably go for langchain/langgraph or llamaindex. they make all the abstractions easier and they allow you to be more flexible about the changes in the industry. about crewai, i love it, but i only use it for internal applications rn. i am not sure it is robust enough yet to power a b2c app (but it is getting there!)
hey can you please edit this tutorial or create a tutorial for accomplishing an actual task with the agents instead of just a simple search? please please edit it to demonstrating the use of the reasoning capacity of the llms
right, you would probably not create this if you are only interested in being a user! this vide is mostly for those who want to understand what goes on under the hood of these kind of technologies or build a product around it :)
@@alejandro_ao Thanks for the answer Alejandro. Ofcourse, this is a learning channel after all :) By the way, video tips: Multi-Agents with LangGraph! Thanks for your work.
hello there! i'm glad that helped! About the course price, totally! I am aware that the pricing is USA/Europe market price range. So you can fill this google form to get financial aid specifying your country and how much you would be able to afford: forms.gle/UjttgZSJnzwLwYJD6 let me know if you have any questions!
Lol it's so wrong xD That does not work, agent will drop your observation if there is something like mass of mercury in it's prompt. You are using tools outside of llm and what it does it's force LLM to use it's general knowlege rather then your observations. Change mass values to some random numbers eg. 1,2,3,4,5,6,7 and see what happens if you want to use Llama 3.1 70B. Tutorial is quite nicly done but this method of passing values won't work in 5 out 10 :p
🔥Join the AI Engineer Bootcamp:
Hey there! The second edition of the AI Engineering Cohort is starting soon 🚀
- Learn with step-by-step lessons and exercises
- Join a community of like-minded and amazing people
- I'll be there to personally answer all your questions 🤓
- The spots are limited since I'll be directly interacting with you
You can join the waitlist now 👉 course.alejandro-ao.com/
Cheers!
Thanks so much for this tutorial. Appreciate going deep inside so we at least have a conceptual view of how these agentic tools work
Beautiful of pure Python - it gives you really deep understanding unlike frameworks! Thank you, man! You have great talent of explaining everything in simple way!
i appreciate it my friend :) take care!
Great video! Please do a video on llama index with librechat.
Excellent video!
You mentioned the JSON format towards the end. Actually, you can ask for a JSON directly in the prompt.
I’ve tried it, and it works really well!”
Example session:
Question: What is the mass of Earth times 2?
Thought: I need to find the mass of Earth
Action: {"action_name": "get_planet_mass", "param": { "planet": "Earth" }}
PAUSE
Can you please share your prompt in this case?
@@gazzalifahim The only difference in the prompt is already mentioned in my comment.
Actually, the credit goes to Hasan Aboul Hasan; this is his example:
ruclips.net/video/cDm5vPXVln8/видео.html&ab_channel=HasanAboulHasan
indeed! you can do this with function-calling LLMs. i am adding this to the next video in the series. was gonna upload it before but had been super busy working on a cohort coruse that i'm doing with some subscribers!
I'm super excited to be considered to be part of your first AI Engineer Cohort Alejandro! Hopefully see you in August. hi from New Zealand!
Hey Jonathan! Thank you! I saw you signed up for the waitlist! Please let me know if you have any questions! 🔥
@@alejandro_ao I am trying to sign up but I am trying to learn more of the curriculum that will be covered in it. Could you provide more details? I couldn't find it on the website
Hi! Thanks for the video, very helpful. Can you explain an example on how to have two or more agents talking to each other? Each one if its own loop
You also use arc? Nice. I was confused as why is my sidebar not closing. Pressed command s multiple times before realising.
This is such an important training! Thank you for sharing brother 🙌🏾💜
hey andy! thank you man! more coming up in this agents series!
Hey! I am going to start teaching and I will be taking a few students.
In short, it is 12 weeks of weekly lectures, exercises and live QAs. We will cover GenAI, LangChain, CrewAI, LangGraph and deployment.
I made a quick landing page to explain what the program is: course.alejandro-ao.com/
I would appreciate your feedback about the landing page, as I am not great at marketing 😅
Thank you so much 🙏
So this works only for mass of the earth but how to make it autonomus for different querys
thanks for very clear and simple explanation
no worries 😎
It would be great to have a video about utilizing ReAct in LlamaIndex after this one. A very advanced one with query rewriting (if possible) and mainly the ReAct in the LlamaIndex. To me, the document load and parse are easy to pick up but ReAct is confusing.
i completely agree with you. query rewriting is one of the best features of llamaindex indeed. that's definitely coming up in this series 😎
@@alejandro_ao I know about LlamaIndex from your videos. I'm very impressed with LlamaParse, especially when they split the document by pages for PDF and presentation files. It's exactly how we consume documents in real life; we read page by page. I mean, it's very natural.
Great explenation. Keep them coming.
yes!
I tried to run the same code with Gemini, but its actually answering all things at once, such as the thought and the action
It would be great if you could orchestrate those agents❤️🙏 using langgraph
i'm working on some videos covering langgraph indeed!
Awesome vid. Really helpful
thank you mate, it's my pleasure!
Alejandro, analyzing the current scenario, do you prefer langgraph/langchain, crewai or without fremeworks?
hey there. for production code, i would probably go for langchain/langgraph or llamaindex. they make all the abstractions easier and they allow you to be more flexible about the changes in the industry. about crewai, i love it, but i only use it for internal applications rn. i am not sure it is robust enough yet to power a b2c app (but it is getting there!)
you can try AutoGen
hey can you please edit this tutorial or create a tutorial for accomplishing an actual task with the agents instead of just a simple search? please please edit it to demonstrating the use of the reasoning capacity of the llms
yes! been very busy recently, but that's the plan 😅
So why do we even need to do this when OpenAI's GPT's already have these internet-search capabilities as well as reading uploaded documents?
right, you would probably not create this if you are only interested in being a user! this vide is mostly for those who want to understand what goes on under the hood of these kind of technologies or build a product around it :)
@@alejandro_ao Thanks for the answer Alejandro. Ofcourse, this is a learning channel after all :) By the way, video tips: Multi-Agents with LangGraph! Thanks for your work.
Hell, yeah brother
can you change the course price as i am from india and cannot afford that much !!
but super video i learnt about some interesting things
hello there! i'm glad that helped! About the course price, totally! I am aware that the pricing is USA/Europe market price range. So you can fill this google form to get financial aid specifying your country and how much you would be able to afford: forms.gle/UjttgZSJnzwLwYJD6
let me know if you have any questions!
@@alejandro_ao thank you sure will fill the form
Thanks @@alejandro_ao, I submitted too
@@rhythamnegi awesome! thank you for your support!
Lol it's so wrong xD That does not work, agent will drop your observation if there is something like mass of mercury in it's prompt. You are using tools outside of llm and what it does it's force LLM to use it's general knowlege rather then your observations. Change mass values to some random numbers eg. 1,2,3,4,5,6,7 and see what happens if you want to use Llama 3.1 70B. Tutorial is quite nicly done but this method of passing values won't work in 5 out 10 :p
Pluto 🥲
i get sad when i think about pluto