I'm just a product manager who knows only a little bit about writing codes, but this video made it really easy to understand the high level concept and get the hang of lang chain. Big shoutout from Japan🍣
this is awesome. as someone who fails with some silly error everytime they try coding, this is the first time i've been able to fluently follow through a tutorial without hiccups. big kudos to you and great work with your explanations. excited to work through your series
I've watched 4 of your videos now, and the "set" and video quality have incrementally improved. I appreciate you putting in the effort to make your videos better. I look forward to watching and learning from your future videos!
Yesterday I finally had a breakthrough and am beginning to understand the things that I see and read. I just hope that I don't have to use API keys as I want EVERYTHING local until I want to access the 'net for more information. I am building a fairly comprehensive application that not only will order groceries but will also perform local actions. What a time to be alive.
Thank you for this really helpful tutorial! It has helped me discover many things to which I was previously unaware of. No more doing things in an amateur way haha!😄
You Good sir are not a Guy who talks the talk. I learnt more about Langchain from you in this half hour than anybody else I have listened to on the last 3months. The veritable quarter to be exact. You are a Guy who walks the walk 🫡
Thanks again, Greg! This video on LangChain concepts was really helpful after watching your LangChain intro. Learning about schemas, models, prompts, etc. is giving me a much better understanding of how LangChain works. Onward to the next video in your playlist!
It has become very difficult to keep up with the ML/DL/AI scene as of lately, so I decided to go with Lang ⛓️, and your video has been the best I've seen so far. Thank you for your effort.
Amazing job explaining the core concepts, this video + the cook book are THE fast references to understand more and memorize less and practice and develop even more. Thanks a million sir
Took me a while to realize the parsing is done by the llm and all you're doing is giving it instructions on how to parse. In hindsight, it's obvious that's what would be done but I'm still amazed and surprised all at once.
Very nice intro, thank you Greg. A good starting point to dig in deeper. Now looking forward to the second part with some use cases and then stop watching videos and get the hands on it. But rest assured, I will sure come back for more videos later. Love your work, please keep it going. Greetings and be well, sir.
This is really helpful. With the order that those concepts are introduced with the great examples, I found everything in the langchain documentation become much easier to follow now. I now know what to look at for each of the ideas I have. Thank you!
I guess we can run it through your cookbook too right? If yes, how would integrating it into my website work? Can we also copy it's code exactly for our initial templates?
@@DataIndependent Hi Greg, I'm a web developer. Recently I tried openai's whisper to do subtitles and I'm amazed by its accuracy. I've also been curious about what Langchain is and how it can used. you offered a great explanation. 🙏
Hi Greg, your content is some of the best around the LangChain library, and I believe you’ll grow a lot in the coming months in the YT tech space. I’ve been studying a lot of this new tools in the past month or so, and experimented with a lot of small exercises. Now I’m really putting it all into test in the real world were I’m trying to create something and I would like to have your feedback on this. I'm building a chatbot that helps my users to get informations about a functionality and execute some actions via API... I was thinking to have the GPT-3.5-turbo ChatAPI as "orchestrator", and if the user wants to get informations redirect the request to a query on a vector DB for getting useful chunks of info and feed those to GPT-4 and get an appropriate response to the user question, and if instead the user wants to execute an action, redirect the request to GPT-4 and the LangChain OpenAPI Agent to execute it and return the result to the user. What do you think about this approach? Any suggestions?
I’ve been doing a similar approach in some smaller projects myself recently. I would definitely recommend trying implementing it yourself first before using langchain to see if you really need it. I made my own chatbot and agent classes that essentially do the same, but much less code. Awesome idea though!
Hey! Just seeing this now - comments got crazy for a while. I would check out LangChain's new conversational retrieval agent which should help blog.langchain.dev/conversational-retrieval-agents/ python.langchain.com/docs/use_cases/question_answering/how_to/conversational_retrieval_agents
Thank you so much for making this so easy to follow and understand. As someone who has been out of the coding game for 15 years, I really struggled with some of the content from others where the assumed knowledge and terminology is so much higher. Keep up the good work :)
@@DataIndependent An app that helps users draft a specific type form of words. I'd like to use an agent that will follow a general process to gather information, then evaluate whether it has enough to draft the text against specific criteria, and ask for more if not. Once it thinks it has enough, it will draft the form of words. Evaluation seems tricky though!
I will write the comment on this video but thanks too much for ALL the videos, code and your explanations. Keep going please, you have talent for this! waiting for new lessons sir!
I know it was just an example of chain, but if we are already using (and spending on) openai gpt, at least on this specific example, we could just have the prompt straight to the full question, which gives the correct answer.
Excellent instruction! You've made what could be a complex topic, very simple. Hope you can do a video on embedding and the various use cases. Thank you for the excellent presentation in this video.
Thanks for great overview. Even as a dev myself I find the docs are dooing a poor job of explaining what is for what and why. You did incredible job at this. Thanks!
Another use case Greg: Imagine I create Wordpress review sites where I review Amazon products. And ai want to review a product category like digital cameras based on the review of the existing cameras. What ai want to langchain to help me review the top 7 cameras, puth the pros and cons of the top 7 cameras and write a review for each camera. The inputs are: product name, amazon link, and I can get the reviews for this product. So I’m thinking about a template like: You are a reviewer of cameras. You found 7 cameras. And you got alle reviews of these cameras. Write an article and give product description, the pros and cons and list the products from 1 to7 based on the reviews.
Thank you for this amazingly helpful overview! Lucky to have this as my intro to GPT world. I have a question regarding splitters and embeddings. I'm working on an application that stores chat history of coaches with their clients and allows quickly find the references from the previous conversations with the user. Let's say he mentioned his dog, so instead of scrolling the chart and trying to find his dog's name and type, I can simply ask GPT and get the name. Would you rather save each message as a different doc/embedding or the whole conversation as one doc?
You explain stuff very well and your voice is quite pleasant. Investing in learning LangChain right now seems to be a wise move for many. It may become something like Docker or Ansible for AI in the future.
Beautiful summary! Thanks a lot for sharing it. I'll definitely check out all the documentation but you gave us a very good overview. Thanks for the ramp up!
If I may make a suggestion, for complete noobs tell them how to or where they can get info on how to run the Python Notebook, that took me a while to figure out, the rest of your stuff was spot on, well done!
As usual, very lucid and high quality content. I think I should embed the youtube transcripts and prompt gpt to 'explain it like data independent'. 😂
Nice! That's fun thank you
Literally amazed at how easily you went through such complex concepts.
Nice and inspiring examples, good job!
I'm just a product manager who knows only a little bit about writing codes, but this video made it really easy to understand the high level concept and get the hang of lang chain.
Big shoutout from Japan🍣
this is awesome. as someone who fails with some silly error everytime they try coding, this is the first time i've been able to fluently follow through a tutorial without hiccups. big kudos to you and great work with your explanations. excited to work through your series
You used the colab notebook to follow the code?
I've watched 4 of your videos now, and the "set" and video quality have incrementally improved. I appreciate you putting in the effort to make your videos better. I look forward to watching and learning from your future videos!
Yesterday I finally had a breakthrough and am beginning to understand the things that I see and read. I just hope that I don't have to use API keys as I want EVERYTHING local until I want to access the 'net for more information. I am building a fairly comprehensive application that not only will order groceries but will also perform local actions. What a time to be alive.
I can't appreciate this video or this playlist more. This work is a masterpiece. Thank you!!
What is abundantly evident is that you, @DataIndependent, are an excellent teacher🙏.
nice! thank you Krbabu that's nice
Thanks a lot for making this! I love that you just went through the notebook, giving us clear and concise overviews of each step.
Wow this is so cool! Love the tip, I hardly get them.
Thank you!
I’m amazed how dense and well indexed this video and document is
Nice! Thank you
One of the best and concise summary on the core concepts of LangChain. I highly recommend it. Thank you.
Thank you for this really helpful tutorial! It has helped me discover many things to which I was previously unaware of. No more doing things in an amateur way haha!😄
Nice! This notebook needs updating forsure
You Good sir are not a Guy who talks the talk. I learnt more about Langchain from you in this half hour than anybody else I have listened to on the last 3months. The veritable quarter to be exact. You are a Guy who walks the walk 🫡
Thanks again, Greg! This video on LangChain concepts was really helpful after watching your LangChain intro. Learning about schemas, models, prompts, etc. is giving me a much better understanding of how LangChain works. Onward to the next video in your playlist!
I have started using Langchain. The video is what I need. Thank you.
It has become very difficult to keep up with the ML/DL/AI scene as of lately, so I decided to go with Lang ⛓️, and your video has been the best I've seen so far. Thank you for your effort.
I had zero knowledge about it and was struggle to understand it. now I have fairly good idea that Langchain is and what it can do with. thanks a lot.
Best langchain explanation I have seen so far. Fast paced. Brilliant.
Where has this channel been all this while? This is gold. Thanks for the great video!
This is super high-quality content. Well done man!
Glad you enjoy it!
Thank you - I am a development editor, and I built a little tool to help ask questions of the first draft text I am sent from what I learnt from you.
Nice!
Hello Rob. Do you have a demo reel of your project?
I am amazed at how well you explained these concepts 🤯Keen to read your newsletters!
Love it! Thank you!
Finally found the clear and intuitive lecture on how to smart use of LLMs by langchain and other search tools. Thank you so much.
Nice! Thank you
High level/big picture explanations like this are very useful to some of us. Thank you
Nice! Glad it worked out
Wow. The power and possibilities are endless! I hooked already.
Your way of explaining is just flawless. Really helpful material, provided in a perfect manner. Congrats!
Nice! Thank you and glad to hear it
Amazing job explaining the core concepts, this video + the cook book are THE fast references to understand more and memorize less and practice and develop even more. Thanks a million sir
Thanks a lot Greg Kamradt for this video, It made me understand very clearly LangChain's coponents.
Took me a while to realize the parsing is done by the llm and all you're doing is giving it instructions on how to parse. In hindsight, it's obvious that's what would be done but I'm still amazed and surprised all at once.
Awesome explanation. So clear! I loved that you just went step by step through the notebook.
Greg, thanks for so generously sharing your knowledge! I like the new navy paint on the walls in your room. 👍🏻
Thank you! it was time for an upgrade
That was a brilliant video. So well described with logical, easily understood examples. Thank you!
Glad it was helpful!
Kudos to you reffort on doing this. Very helpful. Thank you
Too relaxing to learn with you!! The way you communicate is very nice and clear, thank you
Thanks for the kind comments!
Fantastic video. I learned a ton in 60 minutes, by watching this video
Looking forward to watch the rest as well
Nice! Glad to hear it Prasanna!
Very nice intro, thank you Greg. A good starting point to dig in deeper. Now looking forward to the second part with some use cases and then stop watching videos and get the hands on it. But rest assured, I will sure come back for more videos later. Love your work, please keep it going. Greetings and be well, sir.
I love the support! Thank you Markus
In just a few minutes, I became a really big fan! Thank you for your videos!
Nice! Thank you Gabriel
Best video I followed all way long. Thanks Greg. This is Quality content!
Glad you enjoyed it! What're you building
Thank you for this video! You did an amazing job, learning from which we will also do amazing jobs!
This is indeed a Cookbook, very good job, eagerly waiting for the use cases video, thank you!
Glad you liked it!
Brilliant! Would love to see you do one on building a personal assistent with LangChain!
This is really helpful. With the order that those concepts are introduced with the great examples, I found everything in the langchain documentation become much easier to follow now. I now know what to look at for each of the ideas I have. Thank you!
Nice! glad to hear it.
Greg, thanks for another great video. I've come back to this one a few times to clear my head :)
Dude finding this video was one of the best things that happened to me in life
I guess we can run it through your cookbook too right? If yes, how would integrating it into my website work? Can we also copy it's code exactly for our initial templates?
The best explanation I have found on RUclips , thank you!
Awesome thanks Hoyin - what're you working on?
@@DataIndependent Hi Greg, I'm a web developer. Recently I tried openai's whisper to do subtitles and I'm amazed by its accuracy.
I've also been curious about what Langchain is and how it can used. you offered a great explanation. 🙏
All of a sudden, I liked this course. Great content.
BRAVO! Clear, concise, and to the point. Thank you.
This should be a college lecture for all CS students since 2023.
Wow that is an awesome compliment thank you
Okay beta
there won't be a need for a CS degree by 2025...
even in the data science field...
Huh? What's college?
@@greendsnowvery true I didn’t get a degree and I’m working in the CS field. Not easy though
Well crafted overview with concrete examples. I'm very experienced in the field, and this taught me quite a bit.
Great thank you George.
What’re you working on or building?
thanks greg, this was very very easy to understand and insightful
With this attitude your channel will be a start in the upcoming months/years.Keep up the great work..
Nice thank you!
You aced the topic man!. Thanks.
Thank you, I learned so much reading your Cookbook.
Oh heck ya! This is my 2nd tip ever. Love it.
Reach out if you have any questions
Hi Greg, your content is some of the best around the LangChain library, and I believe you’ll grow a lot in the coming months in the YT tech space.
I’ve been studying a lot of this new tools in the past month or so, and experimented with a lot of small exercises.
Now I’m really putting it all into test in the real world were I’m trying to create something and I would like to have your feedback on this.
I'm building a chatbot that helps my users to get informations about a functionality and execute some actions via API... I was thinking to have the GPT-3.5-turbo ChatAPI as "orchestrator", and if the user wants to get informations redirect the request to a query on a vector DB for getting useful chunks of info and feed those to GPT-4 and get an appropriate response to the user question, and if instead the user wants to execute an action, redirect the request to GPT-4 and the LangChain OpenAPI Agent to execute it and return the result to the user.
What do you think about this approach? Any suggestions?
I’ve been doing a similar approach in some smaller projects myself recently. I would definitely recommend trying implementing it yourself first before using langchain to see if you really need it. I made my own chatbot and agent classes that essentially do the same, but much less code. Awesome idea though!
Hey! Just seeing this now - comments got crazy for a while.
I would check out LangChain's new conversational retrieval agent which should help
blog.langchain.dev/conversational-retrieval-agents/
python.langchain.com/docs/use_cases/question_answering/how_to/conversational_retrieval_agents
Great overview Greg! Really enjoyed the examples and the way you broke down the concepts.
Nice!! Thanks man
Great coverage and explanation of Langchain Greg. Thanks for this!
Awesome thank you! What’re you building?
Useful contributions. Thanks your helping the community, Bro!
Nice! Thanks Tim
amazing playlist...watching it completely for sure
Yes! Only tutorial that makes any sense. Great job thank you!!
Awesome! Thanks Mel!
What an amazing video to walk you through the concepts, as well as practical examples. I recommended my friend to watch it too. 😊
Thank you! I’m going to be doing an update soon. Too much code is out of date.
Thank you so much for making this so easy to follow and understand. As someone who has been out of the coding game for 15 years, I really struggled with some of the content from others where the assumed knowledge and terminology is so much higher. Keep up the good work :)
Awesome! Thank you very much - what projects are you working on building?
@@DataIndependent An app that helps users draft a specific type form of words. I'd like to use an agent that will follow a general process to gather information, then evaluate whether it has enough to draft the text against specific criteria, and ask for more if not. Once it thinks it has enough, it will draft the form of words. Evaluation seems tricky though!
This is a great presentation. You have a great way of teaching.
Great, correct, incisive, ultimate pragmatic video explanation, completely zero-based social science students eager to listen
Nice! Thanks
Amazing and great explanation, Ill try out the cookbook in Git. Thank you
I will write the comment on this video but thanks too much for ALL the videos, code and your explanations. Keep going please, you have talent for this! waiting for new lessons sir!
Awesome, thanks you Rocio
Good stuff. Even as a developer the concepts of AI are some completely new so thanks for breaking the concepts down into simple language
Spectacular video. Thank you.
Glad you enjoyed it!
I know it was just an example of chain, but if we are already using (and spending on) openai gpt, at least on this specific example, we could just have the prompt straight to the full question, which gives the correct answer.
Just Awesome .. Thanks a lot for making and sharing this video ..
Dude. Epic💪🏾💪🏾💪🏾💪🏾💪🏾
👏🏾thanks for this Masterclass!
awesome video. the concepts are explained clearly!
Love it thank you Vers.
Very impressive communication skills!
Thanks, this really helped a lot to briefly get an idea what Langchain can do👍
Excellent instruction! You've made what could be a complex topic, very simple. Hope you can do a video on embedding and the various use cases. Thank you for the excellent presentation in this video.
Awesome thank you!
For embeddings, what is the real world use case you want to explore more?
I am gonna set this up to ask questions about langchain to keep me updated with langchain :D
Thank you so much for the video. It was really very helpful. You explained the concepts very well. 🙏
U just blew my mind!!!, jumping into your langchaing guided tour to figure out ways to tame OpenAI 💥
Nice! Thank you
Thanks for great overview. Even as a dev myself I find the docs are dooing a poor job of explaining what is for what and why. You did incredible job at this. Thanks!
This video was a really great beginner overview. Thanks a lot for putting it together. I'm looking forward to part 2.
He got a whole playlist ( 16 episodes ), this one is the 3rd one, you can check it out if you haven't
Awesome introduction about LangChain, great job!
Another use case Greg:
Imagine I create Wordpress review sites where I review Amazon products.
And ai want to review a product category like digital cameras based on the review of the existing cameras.
What ai want to langchain to help me review the top 7 cameras, puth the pros and cons of the top 7 cameras and write a review for each camera.
The inputs are: product name, amazon link, and I can get the reviews for this product.
So I’m thinking about a template like:
You are a reviewer of cameras. You found 7 cameras. And you got alle reviews of these cameras.
Write an article and give product description, the pros and cons and list the products from 1 to7 based on the reviews.
Thank you for this amazingly helpful overview! Lucky to have this as my intro to GPT world.
I have a question regarding splitters and embeddings. I'm working on an application that stores chat history of coaches with their clients and allows quickly find the references from the previous conversations with the user. Let's say he mentioned his dog, so instead of scrolling the chart and trying to find his dog's name and type, I can simply ask GPT and get the name. Would you rather save each message as a different doc/embedding or the whole conversation as one doc?
Insanely high quality video. Thanks so much!
Glad you enjoyed it!
Is this video still up to date? What about the rest of the videos from the series? Anyway, great video. Thanks for helping all of us :)
Thank you for this concise and understandable introduction of the concepts!
Glad it was helpful!
You explain stuff very well and your voice is quite pleasant. Investing in learning LangChain right now seems to be a wise move for many. It may become something like Docker or Ansible for AI in the future.
Thank you Terry. Yes, learning the frameworks that are being put on top of LLMs will be a good investment.
This is a really concise & cool tutorial to start with langchain! Thank you.
Glad it was helpful!
Fantastic presentation! This is incredibly useful. Thank you!
Awesome! Thank you
Thank you for your work Greg! Regards from Belgium :)
Loving the new look bro! Great upgrade and as usual great conent
Nice! Thank you very much. It was time to take AI more seriously.
I'm about to rebrand data indy to my personal brand as well.
Beautiful summary! Thanks a lot for sharing it. I'll definitely check out all the documentation but you gave us a very good overview. Thanks for the ramp up!
Nice glad to hear it
If I may make a suggestion, for complete noobs tell them how to or where they can get info on how to run the Python Notebook, that took me a while to figure out, the rest of your stuff was spot on, well done!
The best overview ever!!
Awesome, thank you!
Thank you very much! This is super helpful for a Langchain Beginner LOL. Looking forward to your use cases!
Thanks Heqing - Working on it
I just f***** love LangChain it is soooo fun
Nice!! What're you building?
Thank you for the guide cookbook! 谢谢你精彩的cookbook！
Awesome! Glad it worked
highly appreciate your work 💖
Awesome thank you
Fantastic tutorial. One of the best I found. Great job! Subscribed
Nice! Thank you
Extremely concise and no hype and straigh to the point. I LOVE IT!
Big thanks for publishing such great content.