How to AUTOMATE LINKEDIN POSTS with CrewAI | CrewAI Tutorial
HTML-код
- Опубликовано: 8 июн 2024
- In this video I'll show you how to use CrewAI, a framework for orchestrating autonomous AI Agents, to automate your LinkedIn posts. We'll describe the basic components of CrewAI and we'll see an example of a custom tool that uses Selenium to download posts from a LinkedIn account. Finally, our crew of agents will generate a LinkedIn post on the differences between Llama 2 and Llama 3.
🤖 Blog post 👉 neural-maze.github.io/blog/po...
🤖 Code 👉 github.com/neural-maze/crewai...
🤖 crewAI 👉 www.crewai.com/
🤖 Llama 3 👉 llama.meta.com/llama3/
🤖 Selenium 👉 www.selenium.dev/
If you want to know a little more about the person behind The Neural Maze ...
------------------------------
🐦 Twitter - / mtrofficus
💻 GitHub - github.com/MichaelisTrofficus
💼 LinkedIn - / moteropedrido
------------------------------
------------------------------
0:00 Introduction
3:05 CrewAI Overview
7:51 Describing my crew
10:23 Github Repo
11:10 Linkedin Scraper Agent
12:10 Custom Selenium Tool
16:10 Web Researcher Agent
17:00 LinkedIn Influencer Agent
17:51 Tasks
19:06 Crew
19:36 Generating a post
21:56 Conclusion
------------------------------
Cyberreality - Mehul Choudhary
/ mehulchoudhary
Creative Commons - Attribution 3.0 Unported - CC BY 3.0
Free Download / Stream: bit.ly/3wVnIXs
Music promoted by Audio Library bit.ly/3TDW3TT
------------------------------
Music ⓒ - Kevin MacLeod - Наука
Finally recruiters are going to talk to me 🚀
On a serious note, great work as usual!
Thaaanks 😁
I watched the video till the end and was sure that you have at least 6 figures in subscribers, but turned out I'm going to be just #242 😄 Very useful and quality content, great job, I'm sure your channel will grow much bigger within few years. 🎉
Thank you very much for your support!! 🙂 I'm very glad that you are finding the content useful 💪🦾
UOUUUUU, Congrats man! Incredible!
Thanksss!!! 😁
Great content! Subscribed
Thanks!! 😀
You'll reach 10k subscribers by the end of the year
Thanks, I hope so! 🙏
top top top, perfeito, obrigado pelo conteudo professor, gataria de saber se tem como fazer isso para ser postado diretamente no instagram e no facebook ?
I guess you could follow the same approach, but changing the tasks to be focused on Instagram / Facebook posts 👍
"Just pay attention to these three concepts: Agent, Task, Tool and Crew". Uh, thats four concepts.
hahaha ups, need to fix that, I added Crew and forgot to update the number 🤣
I installed the requirements.txt file and I got this line finally in my terminal
marketing-crew-py3.11apple@Apples-MacBook-Air crewai_linkedin_post %
I got this error while installing
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
unstructured-client 0.22.0 requires pypdf>=4.0, but you have pypdf 3.17.4 which is incompatible.
Has a virtual environment been created by the code? I have many other projects running with different versions of the programs and I dont want a clash. How do I check if a virtual env has been created? If I have to create one how do I do that?
What does the error mean? WIl it affect the working of the program?
I am quite new to all this and want to learn . So please guide me
Hey!
I recommend you to create a virtual environment, you can use conda if you want, because it makes the work much easier. Also, if you are using PyCharm or VSCode for development, you can also let the IDE generate it for you automatically. Once you have the venv created and activated, just run pip install -r requirements.txt again and it should install the dependencies without any conflicts.
@@TheNeuralMaze Thanks. I will try it out tomorow and let you know if it works fine. I am using VS Code on a macbook. I am NOT using conda.So do I create the virtual env in my project folder, then copy your code from git into the project folder and install the dependencies?
I also want to see a video on how to deploy the crew on streamlit or some similiar online platform, as I want to make it accessible from the web
Yes, that sounds perfect as well. Just create the typical .venv/ in VSCode and install the dependencies from the requirements.txt. I'm thinking about creating some videos about deploying crewAI to streamlit, since Streamlit is a platform that I really enjoy using :)