The Local LLM Crash Course - Build an AI Chatbot in 2 hours! (Udemy.com)

I think it is a great match for me. As a developer with little experience with AI, this course is providing with some hands on practice and educating me in ai tools that were unfamiliar to me. Thank you very much for having it available.

I love it . It was not what I expected. Like someone else has said, there is space for a more in depth analysis of some issues, but overall, as an introduction to what we can do with huggingface and using Lanchain and Chainlit, it's good. For 2 hours, we can't expect a more in depth course. There are some issues with Langchain modules that need some upgrade on the lectures (they changed things), but nothing a good Google search won't solve. Great course.

Absolutely perfect for what I wanted as an intro. I'm not a developer but a sales exec in the space and even though I have not coded in probably 15+ years, I was able to follow along, execute everything, and get a better understanding of the fundamentals behind AI Chatbots ("agents") in just a few hours. Great approach to breaking down the components... I had a blast exploring the available LLMs on Hugging Face... and now I can ask better questions of my company's development and delivery!!! Thank you!

It looks like there have been some changes on Hugging Face since you recorded this video. I can't find the "playing area" shown on your screen, so I might be missing something. I've tried both the Llama and Mistral models, and while it doesn't seem critical, I just want to make sure I'm following the course correctly

That was a concise course, easy to follow, zero clutter, and very helpful resource for getting chatbot foundations right. I recommend to any beginner or anyone who wants to get up to speed with chatbots. Thanks

This is a good introduction course. Fortunately it had a langchain section that it seems was added later, cause without it , the course would be too thin. I would have liked to get more langchain content, but for an introductory course, it was ok.

NOTE: this course is only suitable for LLM beginners! if you are already comfortable with huggingface etc. this course might be not the right one for you I am a starter in the fields of LLM. The course gave me a very good overview of how to start and what possibilities are there. I think i have all the basics i need now to start into my own LLM porject. I also appreciate the extra part about langchain.

This course is too easy and too short, but it is good for understanding the concepts. The instructor doesn’t teach thoroughly, only mentioning most of the important points. If you’re too lazy to search on YouTube, you can take this course. Otherwise, look for more serious courses.

I am a bit disappointed about the scope of the course. The course is more about implementing a private LLM rather than a local LLM. The course teaches how to host an LLM model in Github Codespace (i.e., in the cloud) rather than locally, although a student who is experienced enough will know how to implement locally with the same method. Students are taught just to implement pre-built LLMs, which has little value by itself. Students are not taught to implement an LLM trained with custom data, let alone to train such model. So what I learned here cannot help me do anything meaningful. Now I'm searching for another course for the next step (and I found a few which are at similar price level at Udemy), which probably includes everything taught here. The fact that the LangChain section was added after the course was first launched probably means the scope was not well thought through at first. I'd say if we are just given the codes discussed in the course, we can even save the 2 hours spent on the videos. This is how little I get out of the whole course.

This course took me outside of my python comfort zone. It has given me the private local LLM I wanted. It also improved my python coding technique. Most notably, async functions, decorators, and making my code concise. Thanks!