OpenAI Macro & Micro Strategy: Master Assistants API, Threads, Messages, and Runs
Вставка
- Опубліковано 19 лип 2024
- When OpenAI announces a new set of technology - 1000 startups INSTANTLY vanish. Let's minimize the chances that happens to our products by focusing on embracing GPT technology without overcommitting and by reusable, composable building blocks. Today, we answer a critical question of how to use the Assistants API and more importantly, WHEN to use the Assistants API.
In this video, we delve into GPT-4 Turbo Assistants and their practical applications for builders and engineers. We'll explore a hands-on approach to integrating these technologies into our ongoing project: a Postgres data analytics tool that accelerates the way we interact with our SQL databases. Our focus is on practical, reusable Python structures, offering insights into how GPT-4 Turbo can enhance your product creation process with faster, safer, and more efficient solutions. We discuss the macro and micro implications of OpenAI's latest technologies and how we can stay on the edge of AI product engineering and enhance our personal engineering workflows.
We'll discuss the question of which technologies to focus on - AutoGen, Assistants API, or simple chat completion calls for your projects. Uncover the ideal solution tailored to your specific application needs, be it for automated tasks, context-rich interactions, or quick, straightforward queries.
This series isn't just theoretical - we're building real, applicable tools. In this video we create a reusable GPT-4 Turbo Assistant: TURBO4, and demonstrate its integration with our existing postgres multi-agent system. Our approach is clear and structured, aimed at maximizing the return on your investment in learning and applying these new technologies. We'll provide all necessary code links in the description. If you're intrigued by the potential of conversing with your database or eager to be among the first to test our upcoming 'talk to your database' product, the link will be in the description for that as well. Subscribe, like, and stay tuned for on the ground insights into utilizing the incredible power of GPT-4 Turbo for your engineering and product development endeavors.
👍 THE CODEBASE (Turbo4)
github.com/disler/multi-agent...
🔥 TALK TO YOUR DATABASE
talktoyourdatabase.com
🤖💻 AI Engineering Resources
OpenAI Assistance API: platform.openai.com/docs/assi...
(1) Watch Part One
• Prompt Engineering an ...
(2) Watch Part Two
• One Prompt is NOT enou...
(3) Watch Part Three
• Make AutoGen Consisten...
(4) Watch Part Four
• AutoGen Token Tactics:...
(5) Watch Part Five
• AutoGen SPYWARE: Codin...
(6) Watch Part Six
• Using AUTOGEN & GUIDAN...
📘 Chapters
00:00 OpenAI Just Killed 1k+ Startups
01:35 What we can do on a daily basis
01:00 Recap Our Postgres Data Analytics Tool
05:05 Assistants API: Break it down
06:40 Turbo4 - Assistants API Wrapper Class
10:04 Turbo4 - API Deep Dive
17:14 Replace AutoGen Data Analytics Team with turbo4
20:56 Turbo4 Assistant In Action
24:00 Assistants API vs AutoGen vs Plain LLM
26:23 OpenAI Macro & Micro Strategy
27:30 Macro: OpenAI is becoming the Apple for LLMs
28:40 Micro: Building blocks, Composability, Reusability
29:50 The Ground Is Shifting Beneath Us
30:40 Shift and Move with the Ground
31:37 The Golden Nugget
32:00 Three Videos Left - 7 / 10
🐛 tags
#agentic #gpt #engineering - Наука та технологія
You and Dave Shapiro have the best content. This series has been an emense help in conceptualizing his swarm agent HAAS ideas.
Was thinking the same thing
Well, Dave is great but this guy is a great programmer
@@jaysonp9426 Highly agree.
you Dave and Techfren lol you guys should work together
This series is really quite unique. It's not often that someone builds a project using sound development practices. You have really positioned yourself (this code) to have a 'provider' layer. I can see the implementation of each use case in a single system. Great great! Thanks again for sharing this, love your style and total presentation.
You have the best hands-on videos in the AI space. It's a shame that you have only 6k followers. I started coding in 1984 on a C64, now being a software architect and ATM learning everything I can about AI. Just wanna let you know that I admire your coding philosophy. You are doing all what I consider best practices in coding. The essence of what the GoF design patterns taught us: reusability, composability, separation of concerns, readability, the Unix principle. It's a pleasure to read your code. Chapeau.
Omg I got a late notice but I stopped working as soon as I saw this post lol. This is the video I needed just based off of the title and description.
Not even 3 mins in.. and you already gave me a good idea for my business. I was wondering how I was going to tackle uploading pdfs and having the api spit back the results I can just put that all in a database and query just like here… funny thing is I actually built a natural language query application which is connected to a mysql data base. It outputs the query used and the results .. good stuff man. Thank you.
Open AI is playing real smart. First they released the API and developers built upon this bringing in the users and taking all the risk. Now OpenAI is replicating the in demand features on there own platform. Cheery picking the best products and services.
bro this is awesome. Thank you for all efforts you gave in making this tutorial video, really appreciate it!
This may be my new favorite UA-cam channel
Everything is moving so fast. Always look forward to your videos!
I was so looking forward to your next episode! I’m so glad it’s here!
I was waiting for your video based on the new assistants api.
Great content. Thank you for all the hard work.
Thanks for sharing the codebase, going to run this on wizardcoder and mistral zephyr beta for swarm agents
great job man, this is so valuable! most underrated channel on AI space! Keep up with the great work~
Beautiful tutorial.
I have been super excited for your update and response to all of this!
Thanks so much for the content its really amazing!
Thanks Dan. The winners in this space are going to build the machine that builds the machine. Create software that automates building other software.
100%. This is the endgame mindset.
This is amazing, I am working along with you line by line. One favor, to execute the code would need to setup the postgres with 50 tables, if you could share the dump that would be great help. not sure if you have any "dischord" channel! thanks a lot!
@IndyDevDan First of all, thanks a lot for this amazing series! I am trying to understand a little bit better in which way the OpenAI Assistants API differs conceptually and technically from Autogen. Therefore, I would love to hear your thoughts on this. Is it basically the same idea and almost same approach just packaged in a more accessible way? Is there anything that will only be possible using the OpenAI Agents framework? And most importantly, what was/were the main reason(s) for your shift from Autogen to the Assistants API? Any comment/ replies are highly appreciated 🙏
Just heads up: ua-cam.com/video/KwcrjP3vuy0/v-deo.html This mental model may be changing soon
Take a look at this cookbook comment I noticed:
"Multi-Assistant Thread Runs: single Thread with Messages from multiple Assistants"
This is something I haven't seen anywhere else but... may be exciting if we can have multiple assistants per thread, opens up lot's of possibilities, just letting you guys know.
Loving your content, keep it up.
You are one of the best!
like your code style
Hey guys, is it possible to generate descriptions for images using the assistants knowledge and files/documentation for biased answers for the images description?? Can anyone help me build this up?
I've added few big files like legal codices and in a few interactions token count was over 300k
@Indydevdan Do you think it is still worth me to learn python? Meaning like by the time i do will I be able to just use prompting as higher level language? Please someone let me know
I'm learning it anyway. It's got to help
@@stuartpatterson1617 no question. Are you starting at python or?
Yes. Python is a great language for beginners because it's fairly forgiving and it's also great for experienced coders because for the most part it follows pretty standard syntax that most c type developers are very used to.
Absolutely. Learn Python now and learn it fast. Think of it like stacking building blocks.
If you don't have the foundation blocks the higher levels blocks will fall down.
Python is the (a) foundation, prompt engineering is a higher level brick.
Great series. But I would be VERY careful using the Assistant/Threads it sends everything every request and get quite expensive really fast.
I feel like I’m so far behind. Like what should I do? Should I even bother learning python now?
If you want to work with/in with AI you should. Python is a pretty forgiving language. Excellent for someone just learning to code. 😊
Even if you use AI to generate your code, it's still good to understand what it's doing. So when it doesn't do what you expect, you know how to fix it.
🚀🌈🔥
is this python?
Yes
That being said. Why are you using so many agents to talk to a database? Isn’t this a bit of overkill and costly?
I’m yet to understand what real world products have been built on GPT api it seems like an illusion, I can’t seem to find any evidence of revenue generating products built on GPT
I need sleep
Love this video
🙏
I've been eyeing tools like this. Looks incredible at a glance.