Thank you for the comment. Yeah I am tired of this too, hard to find info that actually helps you build things. Recentl found out that apparently it is very difficult to optimize Autogen Agents because tools like langsmith dont work with Autogen, at least there is no clear implementation. But they work for Crew AI and Langraph, thats why Autogen Agents usually bloated and hell inefficient.
finally 100% correct, so tired off this overhyped youtube`s making everything seem easy when it`s not... great video, thank you
Thank you for the comment. Yeah I am tired of this too, hard to find info that actually helps you build things. Recentl found out that apparently it is very difficult to optimize Autogen Agents because tools like langsmith dont work with Autogen, at least there is no clear implementation. But they work for Crew AI and Langraph, thats why Autogen Agents usually bloated and hell inefficient.
Awesome video! Thanks for sharing all this cool information 😊
No worries, can't wait to make the next video on more Frontline insights from building ai apps for businesses
Were you running your inference on CPU? Not sure why you ran out of server resources.
Me neither, I was looking through logs and it would crash every time I would call it.
welcome back
Thanks :) What do you think about the ROI topic?
The AI makes everything smoother, but I’m concerned about potential security risks with such deep system integration.