crewAI Crash Course For Beginners-How To Create Multi AI Agent For Complex Usecases

CrewAI tools empower agents with capabilities ranging from web searching and data analysis to collaboration and delegating tasks among coworkers. This documentation outlines how to create, integrate, and leverage these tools within the CrewAI framework, including a new focus on collaboration tools.
----------------------------------------------------------------------------------------
Support me by joining membership so that I can upload these kind of videos
-----------------------------------------------------------------------------------

►Data Science Projects:

►Learn In One Tutorials

End To End RAG LLM APP Using LlamaIndex And OpenAI- Indexing And Querying Multiple Pdf's

►Learn In a Week Playlist

---------------------------------------------------------------------------------------------------
My Recording Gear

[gungunvarshney]

Sir sirf apki vajah sa aaj main data science ki field main grow kr paai hu just bcz of you aaj main itna kuch sheek paai hu thank you so much sir for all your countless efforts it means a lot to us

[ravisrivastava]

@krish Naik, you are our Draunacharya, we are your Eklavya ❤🎉😊 love and respect from Stockholm.

[arsalanarsalan]

hey, thanks for informative video, i transitioned from civil to data science 4 years ago, I got duped by coaching institute and i used to go their and watch your videos daily, i must say it helped me in my journey till now.


Regards,
Arsalan

[lixiasong]

Sir, I have researched extensively on this topic, and your content stands out as the best. It offers a clear explanation and an end-to-end demonstration. Thank you so much, please keep the amazing work going!

[raph]

Thanks Krish! I have been waiting for this

[rahulmanocha]

Thanks for getting the requests so seriously 😊

[Georgia-kqsl]

Naya din aur sir ka ek naya course 🎉❤. Sir how are you sooo productive pls tell

[gungunvarshney]

Thank you sir for this wonderful video

[pramodj-vnxp]

Sir, 1. how to pass input variables among the agents?

2. How to have control over final output?
If data is coming from multiple agents like rag, websearch, youtube search, user input etc. I want to store whole data in same file without summarizing what other agents have collected. ( if final agent summarizing i am loosing valuable insights coming from previous agents)
Could you provide insights on it

3. If agent to task is one to one relationship then what is use of having two seperate entities? They could be merged and kept in agent itself with just extra attributes. Why did they choose to keep them seperate?

Look forward to hear from you

[commoncats]

i am waiting when u will start crew happieee😍

[yazanrisheh]

Hi Krish. Can you please do a follow up video on streamlit where we cna also use human input tool so that users can give input they want?

[ShreedharN-rnvm]

Small correction allow delegation is used for hierrarical work load distribution, as you are creating seqential you can use delegation false

[ayushsonu]

Just waiting for this sir thank you very much

[subhankarsahoo]

I think at the end the LLM creation wasnt demonstrated right, did you create the LLM= ChatOpenAI(model= OPENAI_MODEL_NAME)
and then put llm =llm in both the agents as an argument ??
please correct me if i am wrong

[tiyamaheshwari]

Sir since you're using openapi which is a paid service how to use other llms here? can you give source code for huggingface or ollama integration as well cuz when try it, it gives the same error you got at 26:05

[muraliteja]

Hi @krish sir, thank u for sharing knowledge.

[ayaspanigrahi]

quite interesting, how to build custom tools and how we can use custom llms/other llms using hugging face hub

[kirubababu]

Hi Bro,

Could you post a video about the recent multi agent tool Microsoft Magnetic-one tool

[KumR]

Awesome Awesome. The very first thing you mentioned was my biggest doubt so far. Langchain too has an agent so how is crewAI different?

[md.mahadehasan]

I have a question. When we give the topic to the crew, does it need to be exactly the same as any of the video titles of your channel? or is it something like we have to supply something relevant which can be found on the YouTube channel?