Self-Correcting AI Coding Agent + Prompting Deep Dive
2024 ж. 19 Мам.
12 767 Рет қаралды
Self Correcting AI Coding Agent + Prompting Deep Dive
👊 Become a member and get access to GitHub:
/ allaboutai
🤖 AI Engineer Course:
scrimba.com/?ref=allabtai
📧 Join the newsletter:
www.allabtai.com/newsletter/
🌐 My website:
www.allabtai.com
In this video we are gonna dive deep into a self correcting AI coding agent i create. We look deep into how i prompted this system to work pretty good. Using the Clause 3 Opus API + ChatGPT 3.5 Turbo and some custom python code.
00:00 Self-correcting AI Coding Agent Intro
00:18 Flowchart
01:22 Prompting Deep Dive
09:43 AI Coding Agent Testing
17:06 Conclusion
Thank you for putting the time into figuring this out. It gives me a good idea on how to approach something like this! Keep on giving the great content!
Keep it up Kris! Thank you for all you do. Fingers crossed for this contest!
I really enjoy all your creative and innovative AI implementations. Thanks for always sharing the code in your member section.
Insane! Can't believe I'm alive to see AI that can self correct code. Kris is the guy to learn from. I've been a member for a while now and definitely feel that it's worth it!
Kiris, your videos are always so informative! This one was especially helpful, thank you.
Thanks for making this approachable. Really appreciate it!
Another amazing video!Thank you for putting this together and helping us understand all the processes involved
I wanted to ask you if it's possible to create an entire system with the description of a project and the creation of various files for that project just by using a context window of one million tokens. And after creating this complete system with various functionalities, would it be possible to do something like what you did in this video retroactively, improving the system by fixing bugs and adding new features until it becomes exactly as it was written in the initial project? This would be revolutionary and wonderful, and I'd like to know if you could be the first on KZhead to do it
Dude, you have access to different llms. You ask them how to do it. Tell it to break it down into steps. Then break down those steps down because it will likely give you steps that to high level. Anything you don't understand, you ask it to explain it to you at 7th grade level. It's more important for you to learn how to use the tools. Tell it what you want to do and ask it if it's possible. Then tell it to give you an example of what it explained. Also you going to have to go into more detail than just a description of a project. You need to go down to user story level for the system to plan. Then even deeper to write the code.
Let's goo! Amazing content as always! Keep it up Kris
These step by step & high level logic overviews are really helpful.
You're making great content, so much fun trying out there projects!
Keep up the great content Kris! Learning a lot from you:)
I’ve been learning alot thanks man!
Great video! Love the deep dive. Trying it out now!
You're terrific Kriss, keep going with your amazing content, thank you
Great video Kris. I always learn a lot with your experiments. Thanks!
great video! have been looking for this!
awesome content. soaking it in! Thanks for the work
Thank you Kris. Amazing new video as always.
Great workflow. Thanks for sharing
Thank you Kris, Good job as always!
Your videos always make my day. Keep shining!
I;ll definitely try this, it looks quite useful, thanks Kris !
Thank you so much for this!! 😊
absolutely blown away, great job!
You are one of the very few content creators who balance the sensationalism around AI progress with dev based content. Thanks for your efforts
You've got a natural talent for connecting with your audience. Impressive!
thnx mate =)
Glad i found your YT channel, already learned a bunch, thanks :)
:D
I'm addicted to your channel. Can't wait for the next upload!
You rock, Kris! Thanks for your content and experiments.
thnx :)
Great video, I know the snake game has been one of the tests we all try at some stage, but I have never seen it work first time and then iteratively improve. Awesome!
Thank you Kris! loving the channel more Tutorials and walkthroughs please
Great video! Looking forward to implementing agents in my workflow.
Thank you for this video!
Always enjoy the content, and continually, learning
thnx :)
Very interesting. Thank you for this great content!
Interesting approach, gotta try this. Thanks!!
Super cool video on mixing Clause 3 Opus API, ChatGPT 3.5 Turbo, and Python to create a self-correcting AI coding buddy. The walkthrough is insightful and shows some neat tricks.
Thanks Kris! You mentioned vision model to deal with screenshots. And the results were not so good. I recommend to try some OCR API. I tested one of them and it worked great even with photos of monitors.
Great video Kris!
Keep up the good work!
Super video, will try this out 👌🏽👌🏽
This is awesome! Thanks Kris!
Great video, great shirt 😅. Thanks, mate!
Wow again Kris! You mentioned in passing about the video or possibly screen shots overread. I think what you did with Elden Ring combined with this style of agents could really open some amazing use cases. Guided education for new software or games of course. But, possibly a broader automation/RPA method for almost any current software available. Examples: Quickbooks to do your taxes, video editing automation, spreadsheet visualizations. Thx!
Love this channel
Great video.very informative
Really cool, thank you!
It looks interesting - thank you for sharing:)
This was going to be my next project but using local LLM's. Anthropic Opus is really good though, so I may have to use it for code writing and self correction. Nice work!
thnx :)
I can't believe you made this. future of AI coding Agent! wow!
Useful, once again
Great video. Thank you!
This is very interesting and should be fun to explore
thanks for sharing this , Keep it up!
As always great Video 😁
Thanks for quality content
This is pretty awesome!
thnx :)
great content!
Great video!
Great content.
Thank you!
thanks for another useful video
Great project and cool to see how it works. What's the initial setup for this?
Nice content thx
Thanks!
I really like the self correct objective - Is this going up on the repo?
Another great video.
You can make this more efficient by getting the llm to output the code in json. Then deserialise it into a variable. That way we dont need to use a llm to clean the code. { "Code":"" }
will def try that again, struggled with this before
Again I have the feeling to dive deeper in coding
I am curious to know which improvements this code would suggest whilst looking at itself (to have it analyze the code from which it is made) Perhaps it would build a friendly GUI and such .. who knows really. Very useful code solving real issues elegantly! - The compliment! 🖖
thnx a lot, i am gonne try that for sure:P
Fun to watch. How much did this cost you?
Recent member here. Hello!
hey :)
It wasn't bad at all, good video bro.
Kris, thank you for sharing the video. I think the verge website is too easy to crawl (verge is also used in the open Interpreter video). Can you try a more difficult website crawl next time? Thank you
sure! thnx :D
I have a similar setup for running SQL
Let's go! I'm a different name on Discord, btw.
Super fun!
Super! i am ai. Gracias Kris!
Very nice! :)
Pretty darn cool
Can this produce a full GUI or web interface rather than a simple CLI ?
Insane! I am working on a low latency sts and I would love to get some insight on it.
cool :) yeah i think a lot of ppl or working on latecy, i guess if you gonna have a consumer facing product it needs to be low latency, ppl dont have any patience and are used to instant responses
I suppose you are right. As an individual, it is harder to reach a wide audience at the same time but you still pull it off! I think you would be happy to know that I have solved the bugs myself and now adding some spice to the sts assistant :) One thing I wanted to ask was - how do i get my assistant to stop listening as soon as I stop talking ? RIght now I press 'ctrl+c' to interrupt and stop recording.
Thank you for the Video, But how to become member? i see not paid subscription.
Awesome! No LangChain, no Autogen, no CrewAI, just regular-people code!
I’d love the file download for this, tryna implement somthing for personal use this would be great starting point
I don't see any code links, so I won't watch it.
Been needing another option for running and fixing code. Looking forward to the github release of this one.
Autogen
Groq is being a smartass and still adding explanations. Anyone got the proper prompts?
Try with Groq and see how many tries it gets, maybe quantity is better than quality.
more please
This is really cool , although , it is pretty much AUTOGEn but basic, isnt it? Do you see any advantage using this scope for self correcting AI instead of using something like autogen or crew ai?
Greetings from Germany.
Nice
Should try it with a local model
I'm a recent channel member. How do I get access to github?
send me a e-mail :)
how to become a member?
Great
I want something like this for my coding assistant
Wow, cool,😊