Dave thank you so much for this and all of your informative Gen AI videos. Using Instructor, which I found from you, has been one of the most valuable developments I've found. I just purchased your Gen AI launchpad to see some more of these libraries you mention in action, can't wait!
I started building my AI agency and documenting it on my UA-cam earlier this year and I just turned 20 . I’ve built some Ai for gym’s and Accounting firms And this is all thanks to Dave Thank you brother you gave me the confidence to try and start a persona brand. From watching your videos
Awesome video as always! I've used your content before on a project...or two! Could you make a video of some of the real world projects you are building? If its Ok with your clients, and not protected by intellectual property. It might help us to know about some real world use cases.
One strange situation I find myself in: I’m a nurse. I’ve obsessed around software development for the last 2 years. I’m close to intermediate level when it comes to Python and I know how to use most of these libraries/tools. I had to learn about these tools to build some LLM based healthcare use-cases. And you’re telling me software engineers don’t know these frameworks already ? Maybe it’s time for a career change for me or to start something…
1. Awesome video once again, Dave. 2. Finally you mentioned DSPy, giving me an excuse to ask you to make a video about it :3 I'm glad you at least validated it for now - I wasn't sure if I should invest time in getting familiar with it since I still have a bit of trouble getting it. And even when watching hour-long tutorials on it, with people presenting being excited, I still don't get it :/ I figure that if someone's going to make it clear, it's you.
SqlAlchemy and ORMs are anti patterns, they are not needed and we must learn SQL at the end. It's very easy to create a view or a stored procedure to call in the app. Some libraries are too much abstraction with little benefit and some harm, like Langchain, Langfuse, Langsmith, it's easy to do what they do from scratch and be in full control
Can you do a simple AI project with langchain or llama index and implement the same project without these abstractions so we have a better understanding of what you mean? This will be very helpful
can you give an example of why you want to build ai agent system from scratch as oppose to use langchain or langgraph. are they not reliable and optimized?
Thanks🎉. Just a question please: Do you mean that PyMudPDF is working better for some hosted api servics like aws textact? Because i am struggling with extracted pdf chapters from non clear structured files. Some chapters are written as Paragraphs and mixed with some titles? Thanks in advance
Yes, if you run into issues with PyMuPDF or Py2PDF, I would recommend looking into something like Textract or Document Intelligence. You will have much better results with those.
Scikit is learn for training your own models - something a Data Scientist or ML Engineer would focus on. AI engineers focus on integrating pre-trained models.
Hey everyone, I hope you like this overview! What’s missing from this list? Let me know below!👇🏻
The goat of AI engineering content strikes again.
🫡
Your Videos are pure gold. Your practical tips played a huge part in me landing my first client. Veel liefde voor jou uit Duitsland!
I feel ike watching this has furthered my education in agentic AI apps exponentially. You more than earned a sub.
Thank you for the update!
Really respect how you organised your project . Very clean and well thought out. Had to take a screen shot 😂
Dave thank you so much for this and all of your informative Gen AI videos. Using Instructor, which I found from you, has been one of the most valuable developments I've found. I just purchased your Gen AI launchpad to see some more of these libraries you mention in action, can't wait!
Thanks for the support and looking forward to connect in the Discord!
Love this kind of videos, thanks Dave!
You're welcome!
great video, how do you deploy everything?
Damn, so much of libraries to learn on this weekend!
Solid solid solid. Thank you for this.
Great content, thanks for sharing it Dave!
Glad you enjoyed it!
I started building my AI agency and documenting it on my UA-cam earlier this year and I just turned 20 . I’ve built some Ai for gym’s and Accounting firms
And this is all thanks to Dave
Thank you brother you gave me the confidence to try and start a persona brand. From watching your videos
That's amazing! Keep it up 💪🏻
@daveebbelaar thanks so much 🙌🏾
Wow! Great Tutorial on AI libraries definitely worth trying them!
Thanks!
Awesome video as always! I've used your content before on a project...or two! Could you make a video of some of the real world projects you are building? If its Ok with your clients, and not protected by intellectual property. It might help us to know about some real world use cases.
Superb. Thank you
for PDF parser, what do you think of docling? Marker? etc. can you make a video comparing such libraries and suggest when to use them?
Just discovered the channel a couple days, soaking it all in now 😂😂 Thanks Dave for sharing 🫶 And Merry Christmas to you all 🎄🎅
Welcome aboard!
One strange situation I find myself in: I’m a nurse. I’ve obsessed around software development for the last 2 years. I’m close to intermediate level when it comes to Python and I know how to use most of these libraries/tools. I had to learn about these tools to build some LLM based healthcare use-cases.
And you’re telling me software engineers don’t know these frameworks already ? Maybe it’s time for a career change for me or to start something…
1. Awesome video once again, Dave.
2. Finally you mentioned DSPy, giving me an excuse to ask you to make a video about it :3 I'm glad you at least validated it for now - I wasn't sure if I should invest time in getting familiar with it since I still have a bit of trouble getting it. And even when watching hour-long tutorials on it, with people presenting being excited, I still don't get it :/ I figure that if someone's going to make it clear, it's you.
I'll put it on the list! Also a good reason for me to really dive deep into it.
@@daveebbelaar Aweseome
SqlAlchemy and ORMs are anti patterns, they are not needed and we must learn SQL at the end. It's very easy to create a view or a stored procedure to call in the app.
Some libraries are too much abstraction with little benefit and some harm, like Langchain, Langfuse, Langsmith, it's easy to do what they do from scratch and be in full control
Great video ❤
Can you do a simple AI project with langchain or llama index and implement the same project without these abstractions so we have a better understanding of what you mean? This will be very helpful
can you give an example of why you want to build ai agent system from scratch as oppose to use langchain or langgraph. are they not reliable and optimized?
Mirascope as LLM Framework ❤
Can you make a similar video for other roles like ML Researcher and ML Engineer
Thanks🎉. Just a question please: Do you mean that PyMudPDF is working better for some hosted api servics like aws textact?
Because i am struggling with extracted pdf chapters from non clear structured files. Some chapters are written as Paragraphs and mixed with some titles? Thanks in advance
Yes, if you run into issues with PyMuPDF or Py2PDF, I would recommend looking into something like Textract or Document Intelligence. You will have much better results with those.
Hi @daveebbelaar can you please make a video on celery for AI products
Libraries are not important insight is, if you have insight you can learn these libs easily.
thx frenz
But no Scikit in there??
Scikit is learn for training your own models - something a Data Scientist or ML Engineer would focus on. AI engineers focus on integrating pre-trained models.
Pydantic AI? I think it's way better than Langchain and Llamaindex
you spelled "C programmers did all the work for you" wrong.
...make the transition to...
accc