Thank you so much! I have been struggling to make a simple chatbot using the Assistent API. No problem to get the first response, but it was really difficult to get the next. A modified version of your script solved the issue. Such a good presentation and demonstration 🙂
Nice walkthrough! I think one major advantage you did not go deep into is that the assistants API seamlessly integrates code interpreter, which has a lot advantages in certain domains. However, I also agree that the customizability from chat completions offer a more controllable RAG solution. I am currently considering trying to implement an assistant API function with code interpreter which I can utilize through a function call within the chat completion - thoughts on this approach? Additionally, I think building agent swarms with the assistants API is interesting - would love to see a video from you on this topic;) Keep up the good work!
Thank you so much for the video. This was the most helpful video I have watched on this subject so far. I immediately went and signed up for Data Alchemy and I am very impressed by the offerings there. Thanks again!
Thank you for this video!! You answered my big question: - ‘should i go all in?’ because i also have a current codebase linking openai, pinecone, etc. And, i like having that control :)
Thanks so much for this video. I would still build a project on this while keeping an eye on it's development as time goes on. Again, thanks so much for the video. Kudos to you!
i share the same hesitations as you do too. I think the downsides are the same as OpenAI 3.5 fine tuning, lack of control being the key. Ease of access is great, and I really wanted to use this for simple personal apps as the threads management is ideal for memory management, but lack of streaming and control is a tough sell.
Just to note, unless you have $10 credit in your openai account, you can't use the model listed in the video. You can use: model="gpt-3.5-turbo-1106" for free and it works for the retrieval tool. Also Dave I think you forgot to add the FAQ pdf in the repo? Cheers though, nice to set this up last night, will wait for the rest 👍
Thanks Dave. Great content as always, what do you think about the retrival quality. You mentioned in 23:51 that that you had some issues. How would you rate the overall experience on a scale of 10? Clearly build this block by block usiong other framework takes a lot more work. And retrival time sometimes is quite long. Would be awesome if you can do another tutorial on this topic!
Hey Dave, love your videos! You're really an inspiration to someone studying in the same field (AI). I'm hoping to also do some freelance work when I get a bit further into my studies. Can I ask how you managed to acquire clients when you first started? Has any of the connections been made through RUclips?
That's awesome! When you're starting out, I would recommend using your network and LinkedIn. I now get clients through RUclips, but this was not the case for the first 4 years of my freelance career.
🎯 Key Takeaways for quick navigation: 21:21 🧠 *Understanding OpenAI Assistant Integration* - Overview of integrating OpenAI Assistant with user-specific queries. - Demonstration of handling different user queries for John and Sarah. - Mention of the simplicity of achieving this in about 100 lines of code. 22:05 🚀 *Should You Go All-In on OpenAI Assistance API?* - Discussion on the hesitation to convert existing applications to OpenAI's new API. - Current approach of building custom functionality using length chain Vector databases. - Considerations about data control, beta version concerns, and potential issues with API responses. 24:37 🧭 *Future Recommendations and Exploration* - Encouragement to understand OpenAI Assistant concepts like managing threats and messages. - Recommendation to explore and play with the Assistant API while being cautious for full-scale applications. - Mention of the Alchemy Codex and free group "Data Alchemy" for further learning. 25:50 📚 *Closing Remarks and Future Plans* - Gratitude for watching and a request to like and subscribe. - Apology for recent absence, explanation of personal changes, and commitment to consistent future uploads. - Teaser for upcoming content, including the complete WhatsApp Bot video. Made with HARPA AI
Thanks for the video I have two questions. 1)Is there any difference between assistent and custom gpts ? 2)and How can i integrate them in instagram or whatsapp ?
Did everything exactly, but the assistant is not being created. You have an error in the code in line 34 (the file variable is undefined), but somehow the code was compiled anyway. How can I fix the error?
hey, great content keep it going. i would like to know how do you split your screen like that and on the right side you have some informations about code being executed etc. would love to learn that, thank you very much in advanced
I was hoping you would show a way to output the chat to Whatsapp, so far I managed to do it with manychats, which is actually convenient also to build the workflow visually and not code it, as well as using built in function like forms and canned responses, but all in all if there was a way to do that without proprietary software it would be even better.
I will cover that in a next video! You can check out the repo already that is linked in the description. All the code and instructions are there already ;)
, a question, im new to OAI assistant API..i have already had 'credentials' from my custom GPT in GPT store. how can i connect it to my separate app with assistant API? not recreating the instruction on assistant API but just connect it to custom GPT in GPT store that i have already created earlier
Hi Dave, loved the video! Talking about RAGs, is it possible to create an assistant that would do NLP to SQL and connect directly to a database (like Snowflake etc.)? Although Langchain does provide an SQLDatabaseChain to do this exact same thing, but it's inefficient and can't track older messages - so I was thinking this could be a good assistant. But how, is my question.. how would this assistant create and test those queries in a database? Any ideas? TIA :)
@@daveebbelaarhope you are creating another video for this WhatsApp OpenAI assistant! Very interesting use case. Your tutorials are the best out there! 👍🏽
Generated by Talkbud: 📝 Summary of Key Points: 📌 The video tutorial explains how to work with the OpenAI Assistance API to build your own bots. It covers creating an assistant, enabling retrieval, uploading data, generating responses, and running the assistant. 🧐 The concept of threads and messages is introduced, allowing developers to keep track of different conversations. The tutorial demonstrates how to create and store threads based on unique identifiers and discusses using a database solution to store and retrieve thread information. 🚀 The speaker recommends keeping an eye on the development of the Assistance API and mentions their free online community, Data Alchemy, where they teach data science and artificial intelligence. 💡 Additional Insights and Observations: 💬 "The Assistance API is a powerful tool for building conversational AI bots." 📊 No specific data or statistics were mentioned in the video. 🌐 The speaker mentions their free online community, Data Alchemy, as a resource for learning data science and artificial intelligence. 📣 Concluding Remarks: The video tutorial provides a step-by-step guide on using the OpenAI Assistance API to build bots. It covers the creation of an assistant, retrieval, data uploading, generating responses, and running the assistant. The concept of threads and messages is explained, along with the use of a database solution for storing and retrieving thread information. The speaker recommends keeping an eye on the Assistance API's development and mentions their free online community, Data Alchemy.
There's so much assumed knowledge here that "true" beginners really get lost. You literally need to start from the basics to educate people like me who have no prior knowledge of this.
>> I believe we are meant to be like Jesus in our hearts and not in our flesh. But be careful of AI, for it is just our flesh and that is it. It knows only things of the flesh (our fleshly desires) and cannot comprehend things of the spirit such as peace of heart (which comes from obeying God's Word). Whereas we are a spirit and we have a soul but live in the body (in the flesh). When you go to bed it is your flesh that sleeps but your spirit never sleeps (otherwise you have died physically) that is why you have dreams. More so, true love that endures and last is a thing of the heart (when I say 'heart', I mean 'spirit'). But fake love, pretentious love, love with expectations, love for classic reasons, love for material reasons and love for selfish reasons that is a thing of our flesh. In the beginning God said let us make man in our own image, according to our likeness. Take note, God is Spirit and God is Love. As Love He is the source of it. We also know that God is Omnipotent, for He creates out of nothing and He has no beginning and has no end. That means, our love is but a shadow of God's Love. True love looks around to see who is in need of your help, your smile, your possessions, your money, your strength, your quality time. Love forgives and forgets. Love wants for others what it wants for itself. Take note, true love works in conjunction with other spiritual forces such as patience and faith (in the finished work of our Lord and Savior, Jesus Christ, rather than in what man has done such as science, technology and organizations which won't last forever). To avoid sin and error which leads to the death of our body and also our spirit in hell fire, we should let the Word of God be the standard of our lives not AI. If not, God will let us face AI on our own and it will cast the truth down to the ground, it will be the cause of so much destruction like never seen before, it will deceive many and take many captive in order to enslave them into worshipping it and abiding in lawlessness. We can only destroy ourselves but with God all things are possible. God knows us better because He is our Creater and He knows our beginning and our end. Our prove text is taken from the book of John 5:31-44, 2 Thessalonians 2:1-12, Daniel 7-9, Revelation 13-15, Matthew 24-25 and Luke 21. Let us watch and pray... God bless you as you share this message to others.
🔐 Copy my AI tools & workflows: bit.ly/data-alchemy
💰 Kick-start your data freelance career: www.datalumina.com/data-freelancer
Thank you so much! I have been struggling to make a simple chatbot using the Assistent API. No problem to get the first response, but it was really difficult to get the next. A modified version of your script solved the issue. Such a good presentation and demonstration 🙂
hey could u help me out im gettin an error
Nice walkthrough! I think one major advantage you did not go deep into is that the assistants API seamlessly integrates code interpreter, which has a lot advantages in certain domains. However, I also agree that the customizability from chat completions offer a more controllable RAG solution. I am currently considering trying to implement an assistant API function with code interpreter which I can utilize through a function call within the chat completion - thoughts on this approach? Additionally, I think building agent swarms with the assistants API is interesting - would love to see a video from you on this topic;) Keep up the good work!
Will definitely cover the code interpreter and function calling in future videos. The idea of these "swarms" also intrigues me.
So good to see you Dave, love your work, gonna follow this tutorial def!!
Good to see you here again as well Peter!
Awesome video and teaching style, thanks!
Dude, this is really cool😄
Congrats man!
Great work Man, Thank you
Thank you so much for the video. This was the most helpful video I have watched on this subject so far. I immediately went and signed up for Data Alchemy and I am very impressed by the offerings there. Thanks again!
Thank you for this video!! You answered my big question: - ‘should i go all in?’ because i also have a current codebase linking openai, pinecone, etc. And, i like having that control :)
Thanks so much for this video. I would still build a project on this while keeping an eye on it's development as time goes on. Again, thanks so much for the video. Kudos to you!
Thanks!
you deserved all the likes, thank you
i share the same hesitations as you do too. I think the downsides are the same as OpenAI 3.5 fine tuning, lack of control being the key. Ease of access is great, and I really wanted to use this for simple personal apps as the threads management is ideal for memory management, but lack of streaming and control is a tough sell.
Great Tutorial , Thank You so much for this
This really is awesome content Dave. I really enjoy the video
Just to note, unless you have $10 credit in your openai account, you can't use the model listed in the video. You can use: model="gpt-3.5-turbo-1106" for free and it works for the retrieval tool.
Also Dave I think you forgot to add the FAQ pdf in the repo?
Cheers though, nice to set this up last night, will wait for the rest 👍
Thanks for pointing that out! The /data/ was in the .gitignore file. I just fixed it ;)
Great video pal
Thanks! 🙏🏻
Who needs netflix when you can binge watch this…
thanks so much, this allows for so much development potential!
Absolutely!
Please consider increasing font size of the code editor when recording videos. Thanks for the great content.
Thanks Dave. Great content as always, what do you think about the retrival quality. You mentioned in 23:51 that that you had some issues. How would you rate the overall experience on a scale of 10? Clearly build this block by block usiong other framework takes a lot more work. And retrival time sometimes is quite long. Would be awesome if you can do another tutorial on this topic!
Hey Dave, love your videos! You're really an inspiration to someone studying in the same field (AI). I'm hoping to also do some freelance work when I get a bit further into my studies. Can I ask how you managed to acquire clients when you first started? Has any of the connections been made through RUclips?
That's awesome! When you're starting out, I would recommend using your network and LinkedIn. I now get clients through RUclips, but this was not the case for the first 4 years of my freelance career.
🎯 Key Takeaways for quick navigation:
21:21 🧠 *Understanding OpenAI Assistant Integration*
- Overview of integrating OpenAI Assistant with user-specific queries.
- Demonstration of handling different user queries for John and Sarah.
- Mention of the simplicity of achieving this in about 100 lines of code.
22:05 🚀 *Should You Go All-In on OpenAI Assistance API?*
- Discussion on the hesitation to convert existing applications to OpenAI's new API.
- Current approach of building custom functionality using length chain Vector databases.
- Considerations about data control, beta version concerns, and potential issues with API responses.
24:37 🧭 *Future Recommendations and Exploration*
- Encouragement to understand OpenAI Assistant concepts like managing threats and messages.
- Recommendation to explore and play with the Assistant API while being cautious for full-scale applications.
- Mention of the Alchemy Codex and free group "Data Alchemy" for further learning.
25:50 📚 *Closing Remarks and Future Plans*
- Gratitude for watching and a request to like and subscribe.
- Apology for recent absence, explanation of personal changes, and commitment to consistent future uploads.
- Teaser for upcoming content, including the complete WhatsApp Bot video.
Made with HARPA AI
Good stuff 🙌
🤝
Great tutorial! Dave, what's the extention you used to run python in real time on the right?
Thanks! Check out this video: ruclips.net/video/zulGMYg0v6U/видео.html
Thank you for the video - very usuful. The new APIs are great.
Glad it was helpful!
Thank you for this video! Can you consider recording function calling without using langchain?
Thanks for the video
I have two questions.
1)Is there any difference between assistent and custom gpts ?
2)and How can i integrate them in instagram or whatsapp ?
Great tutorial!
How can we connect it to a library like streamlit to be able to give non developer users to play with my assistant in web env?
Nice Video, great to hear you are also finding the same issues with the BETA version of the Assistant 😂
Did everything exactly, but the assistant is not being created. You have an error in the code in line 34 (the file variable is undefined), but somehow the code was compiled anyway.
How can I fix the error?
make video in darkmode please.
great work
thank you far this amazing vid
You're very welcome!
hey, great content keep it going. i would like to know how do you split your screen like that and on the right side you have some informations about code being executed etc. would love to learn that, thank you very much in advanced
Thanks! Check out this video: ruclips.net/video/zulGMYg0v6U/видео.html
I was hoping you would show a way to output the chat to Whatsapp, so far I managed to do it with manychats, which is actually convenient also to build the workflow visually and not code it, as well as using built in function like forms and canned responses, but all in all if there was a way to do that without proprietary software it would be even better.
I will cover that in a next video! You can check out the repo already that is linked in the description. All the code and instructions are there already ;)
@@daveebbelaar Thanks a lot, keep up the good work!
, a question, im new to OAI assistant API..i have already had 'credentials' from my custom GPT in GPT store. how can i connect it to my separate app with assistant API? not recreating the instruction on assistant API but just connect it to custom GPT in GPT store that i have already created earlier
Hey Dave! Do we need to buy ChatGPT Plus for this?
Please please please reply 🥺
no, I used my API credits only
Hi Dave, loved the video!
Talking about RAGs, is it possible to create an assistant that would do NLP to SQL and connect directly to a database (like Snowflake etc.)?
Although Langchain does provide an SQLDatabaseChain to do this exact same thing, but it's inefficient and can't track older messages - so I was thinking this could be a good assistant.
But how, is my question.. how would this assistant create and test those queries in a database?
Any ideas?
TIA :)
Great video, thanks! Do you take on private clients to implement the above and other projects?
Yes! Check out Datalumina's website in the description.
Can you do a video explaining the function calls and incorporating it into a bot tutorial like this? Plz I'm so confused 😭
Noted!
Do we need premium ChatGPT model to use OpenAI 's "Retrieval" tool for creating personal Assistant?
no, I used my API credits only
Can you make screen more bigger please
I didn't understand how is this related to airbnb and whatsapp? how does the bot sends messages to whatsapp?
That’s for another video. This part only covered the OpenAI integration. Not the WhatsApp integration.
@@daveebbelaarhope you are creating another video for this WhatsApp OpenAI assistant! Very interesting use case. Your tutorials are the best out there! 👍🏽
Generated by Talkbud:
📝 Summary of Key Points:
📌 The video tutorial explains how to work with the OpenAI Assistance API to build your own bots. It covers creating an assistant, enabling retrieval, uploading data, generating responses, and running the assistant.
🧐 The concept of threads and messages is introduced, allowing developers to keep track of different conversations. The tutorial demonstrates how to create and store threads based on unique identifiers and discusses using a database solution to store and retrieve thread information.
🚀 The speaker recommends keeping an eye on the development of the Assistance API and mentions their free online community, Data Alchemy, where they teach data science and artificial intelligence.
💡 Additional Insights and Observations:
💬 "The Assistance API is a powerful tool for building conversational AI bots."
📊 No specific data or statistics were mentioned in the video.
🌐 The speaker mentions their free online community, Data Alchemy, as a resource for learning data science and artificial intelligence.
📣 Concluding Remarks:
The video tutorial provides a step-by-step guide on using the OpenAI Assistance API to build bots. It covers the creation of an assistant, retrieval, data uploading, generating responses, and running the assistant. The concept of threads and messages is explained, along with the use of a database solution for storing and retrieving thread information. The speaker recommends keeping an eye on the Assistance API's development and mentions their free online community, Data Alchemy.
What’s Arc? Is that an app?
Web browser
What I’ve been struggling with is to connect the assistant api to my slackbot
Is that free to create open ai assistant
No, you need credit balance to get an API key.
What if you arent a developer and you want to do this?
Then I would look at tools like Botpress and Voiceflow.
Can you share your code?
The Github URL is in the description.
I think langchain is still easier
Agree, and you have way more control.
@@daveebbelaar do you find a vector store better than others? Building an app with chroma but not sure if good choice
I wouldn't exactly call this for "beginners"...most beginners don't run python lol...great info though
Question, how did he store the pdf file as a path variable?
There's so much assumed knowledge here that "true" beginners really get lost. You literally need to start from the basics to educate people like me who have no prior knowledge of this.
Or you start investigating on your own? Write what you dont know and ask GPT?
>> I believe we are meant to be like Jesus in our hearts and not in our flesh. But be careful of AI, for it is just our flesh and that is it. It knows only things of the flesh (our fleshly desires) and cannot comprehend things of the spirit such as peace of heart (which comes from obeying God's Word). Whereas we are a spirit and we have a soul but live in the body (in the flesh). When you go to bed it is your flesh that sleeps but your spirit never sleeps (otherwise you have died physically) that is why you have dreams. More so, true love that endures and last is a thing of the heart (when I say 'heart', I mean 'spirit'). But fake love, pretentious love, love with expectations, love for classic reasons, love for material reasons and love for selfish reasons that is a thing of our flesh. In the beginning God said let us make man in our own image, according to our likeness. Take note, God is Spirit and God is Love. As Love He is the source of it. We also know that God is Omnipotent, for He creates out of nothing and He has no beginning and has no end. That means, our love is but a shadow of God's Love. True love looks around to see who is in need of your help, your smile, your possessions, your money, your strength, your quality time. Love forgives and forgets. Love wants for others what it wants for itself. Take note, true love works in conjunction with other spiritual forces such as patience and faith (in the finished work of our Lord and Savior, Jesus Christ, rather than in what man has done such as science, technology and organizations which won't last forever). To avoid sin and error which leads to the death of our body and also our spirit in hell fire, we should let the Word of God be the standard of our lives not AI. If not, God will let us face AI on our own and it will cast the truth down to the ground, it will be the cause of so much destruction like never seen before, it will deceive many and take many captive in order to enslave them into worshipping it and abiding in lawlessness. We can only destroy ourselves but with God all things are possible. God knows us better because He is our Creater and He knows our beginning and our end. Our prove text is taken from the book of John 5:31-44, 2 Thessalonians 2:1-12, Daniel 7-9, Revelation 13-15, Matthew 24-25 and Luke 21. Let us watch and pray... God bless you as you share this message to others.
Thank you so much for this video!