Let`s Build AI: MEGA AUTOCOMPLETER (LLM Powered) | AI Engineer Project
Вставка
- Опубліковано 12 лип 2024
- Let`s Build AI: MEGA AUTOCOMPLETER (LLM Powered) | AI Engineer Project
👊 Become a member and get access to GitHub and Code:
/ allaboutai
🤖 Great AI Engineer Course:
scrimba.com/learn/aiengineer?...
🔥 Open GitHub Repos:
github.com/AllAboutAI-YT/easy...
📧 Join the newsletter:
www.allabtai.com/newsletter/
🌐 My website:
www.allabtai.com
Lets build a project together using LLMs. This project is an LLM powered Autocompleter with a functional UI powered by React and Node.js. Bulding an app with LLMs in a fast simple way.
00:00 Project Intro
01:52 Prompt Setup
06:02 Running the prompt
06:34 Lets Build
10:42 Itterations
13:18 Adding a new feature
16:58 Fixing a bug
18:15 Final App - Наука та технологія
As a front-end web dev of 25 years (html, css, js, react, python), I would recommend using Flask as an API for your Python and build out the UI separately. This would allow you to continue to build out features with Python (local LLM, RAG, ...) on the back-end and allow any front-end (from static html/css/js, to react, to mobile apps...) to be used. You can then leverage everything you've already done in Python FAR more easily :)
I've been doing this - starting with your local RAG/vault demo, and it's been working out very well!
several hours for completely developed a system (for everyone who knows prompting and logical thinking!)....what a world it will become!!
youre a genius bro
I always enjoy your videos! Thanks!
As a dyslexic mind that mastered the art of its strengths and weaknesses, l normally utilize a critical reliance prompting techniques, spatial intelligence, critical thinking with problem solving skillset and so on... I love Human General Intelligence collaborating with AI knowledge Application capabilities...
This was great, seeing how you can use gpt to boost your productivity.
Thank you - great video:) Usualy I use streamlit as front and python as back - the simpler solution for me:)
you have autocompletion in the vscode terminal by what method?
Have you tried a combination Chain of thought and chain of action
you should do this again but use python for the backend.
My front end game is pretty much non-existent. By the way, what is the cost of the OpenAI API key?
The API key (actually they are switching to project keys) is free but you have to pay for the tokens you use. About 1 million tokens cost about $17 and would be sufficient for about 750k words. For scale: LoTR has around 500k words (all books).
I don't exactly know how it would measure words in source code though.