Видео

Wow! Wonderful... This is web design simplified and summarized in 10 minutes! What I couldn't learn from 100 videos on RUclips, I have just learned in a 10 minute single video. That's awesome and marvelous! Thanks very much for this wonderful update. No need to ask if I have subscribed to this channel. 🙆

My AI are my co-worker. I just gave him some options to approach a problem and what it thinks is the the best course of action. I am not really good at making choices so I gave that part to the AI. It won't always go right, so I just talk to him and whatever. Like a real person lol idk

something i learned while taking my first class in Basic that i didnt realize i needed to apply in c++ write your code in psuedocode then write out your code on paper and see if it matches. no reason i got an A+ in BASIC a C+ on C++

I disagree entirely, this video is too dogmatic. If your options are to lean too heavily on the crutch of AI tools for building a shitty version of a project you are excited about, or give up on coding. The former is clearly the better choice. There are ways to use it, there are ways to abuse it.

Making yourself dependant on all these AI tools is the biggest mistake you can make as a dev. Learning by failing over and over again is how you actually learn. You should always ask yourself: "What good can I build with only a PC?" No internet, no fancy formatters, no syntax highlighting, no snippets, no autocomplete just you and the machine. If you can build shit like that, I'll actually respect you as a dev. Not by how fast you can build some AI generated bullshit.

I see it differently: not everyone is a nerd and wants to learn an abstract programming language or several of them. Some people are extremely bored by technical details, but maybe they are a good UX designer with a great app idea that they have always wanted to implement but couldn't because all this technical mumbo jumbo annoys them. Besides, complex projects can take months or even years to develop without making any money or enough money. Not everyone has the same aspiration and wants to learn the same thing. These people would never be able to realize their ideas because they lack the skills to do so or don't have the money to hire expensive developers who, out of laziness or incompetence or lack of experience, simply tell you that something is not technically feasible and then you get a mediocre implementation with many compromises. Software is so much more than just programming, I would even say that ideas and a great UX design are much more important than being able to program something yourself. If I want to build a house, it's perfectly OK for me to buy ready-made parts at the hardware store and not to make the front door myself or to weave the carpet. I am still proud of my achievement after completion because the idea and the floor plan came from me, even if an architect helped with it, because I know nothing about statics and architectural drawings. It depends on what you want to achieve, but I wouldn't be the type to work on my motorcycle on Sundays either. There are also an enormous number of brilliant musicians who have composers, studio technicians and sound engineers at their disposal in the studio, but the ideas for that brilliant hit come from them and they sing the song. Should they not be proud of themselves just because they don't know how to operate a mixing desk? Not everyone gets their satisfaction from the same things. AI democratizes creative expression.

nice bro you helped me

You Are Great Bro !

I've found ai to be a fantastic teacher and reviewer. It's especially helpful for things that you don't know you don't know. I initially used it to do everything for me, like a script kiddie, but it ended up failing most of the time. It was taking on too much at once, creating bugs in large chunks of hard-to-debug code, or simply did not understand the rest of the project. Ai isn't at the point where it can truly understand your project, making it your job to do. At least the ai I've been using. It'll probably end up doing a better job than humans eventually, but I don't think it's quite there yet.

ask chatgpt to make ascii in code💀💀💀

yes elementor has these responsive options for width and height of elements. thanks for the video

if this is what you understand from ai you dumb. Why to learn code if ai will do it faster and efficient? Ai is a tool to speed up everything and it will be in many areas

This video is better than all RUclips video related code , best for exact direction and discipline.. thanks a lot ❤

so much knowledge

1) Get hooked on blockchain and join a Discord full of insanely talented devs. 2) Dive into GitHub, break down bots, and slap pieces of code together to make sense of it all. 3) Struggle endlessly over basic scripts (like listening to BSC transactions) but slowly get somewhere. 4) Discover ChatGPT and start offloading the boring parts, using the extra time to debug and improve. 5) **Feel guilty about not writing every line of code myself anymore, but realize it’s part of learning.** 6) Let go of the guilt, aim higher, and take on even more challenging projects.

90% of the time i prefer to write the code on my own and i literally detest the chatGPT code but here is the catch is when to use it? I use it for the redundant tasks and detecting where is the bug but i solve them myself, it can help you in a project minimize the build time so i use it for the UI boring optimizations but the logic is mathematically complex so i prefer to do it myself ChatGPT can be useful only when you know to learn and use it correctly otherwise if you use it for everything then you are not a software engineer Sorry my English is bad

got a new subscriber

copypasting code = being able to copy paste, not being able to code or program, anyone can call themself a builder for using ai but when it comes to doing it themselves, well..

I was talking with someone a couple of weeks ago who was saying I should use AI to help me learn Python (I'm an FE dev). I tried to explain to him that AI can't teach you anything and there are some things a person just has to do by themselves. ...I really don't think they got it.

Whenever i start a new project i only use ai to create a roadmap and to help me a bit along the way in smaller tasks. If AI will be always available whats the harm in using it?

I used ChatGPT, Copilot, and Amazon Q to build a webapp hosted on AWS amplify using the same chats for each one over the course of a couple days It uses React/Typescript with GraphQL to populate my dynamoDB from the metadata from the files uploaded to a specific s3 and use that data to populate a separate field on another page in an organized chart and has a nice clean look to it. It got the job done but years ago I wanted to be a developer and I used to write my own stuff and troubleshoot for hours on end and even though the job got done I hardly learned a thing if there was something I wanted to go back and add or change I would probably be clueless of what to do and using chatgpt over the last year has made me forget pretty much most of what I knew I find myself even having to re-learn css and html at times. I have been working with code and in IT for years enough to troubleshoot basic concepts but I have also come to realize alot of the AI is really dumb and I find myself still having to research and correct the AI to get it to do what I want. Moral is if you want to seriously be a developer chatgpt is ruining your ability to critically think on your own and making people lazy. The only experience I got out of it though was the AWS knowledge of configuring everything for it to work but when you don't even know what is in your code because you didn't type it that became a huge task on it's own.

I'm currently learning MERN and it's so hard because I have to remember all these functions etc, Even on just the basic beginner project of a product database app. My approach is deleting my folder completely and trying to recreate it as it was before, while asking AI why certain stuff was done. I did this about 6 times now and am at the API stage, but it's a bit difficult to remember all the details. Having the VS Code Blackbox AI extension helps, but I try not to rely on it too much, and I am contemplating simply disabling it till I get the basics down fully. At the moment, I feel a bit more confident knowing my skills are growing and that I'm getting better daily, but it's a bit rough since I wanted to become a frontend developer solely. On a good note, I can now understand exactly what the tutorials I watch are doing in terms of the setup, even from different RUclipsrs. At first, it was like a jumbled mess of random code. I hope to refine my skills even further so I can create something impactful!

When "fuck around and find out" is a good thing

i use gpt for simple stuff not my backend.

I’m two years out of college and currently working as a Head of Engineering at a startup. Let me share a perspective that might resonate with those in similar situations: people like me are in a prime position to leverage AI, especially LLMs like ChatGPT, to their maximum potential-and here’s why. When ChatGPT-3 launched during my senior year, it was revolutionary. Everyone was amazed at what it could do. However, the reactions from different groups were telling: Mid- to senior-level engineers quickly tested its limits and dismissed it as "useless" because it couldn’t match their nuanced expertise. Beginner programmers leaned on it completely, relying on it as a crutch while failing to grasp its limitations or the fundamentals of programming. But for someone like me-who had worked on 10+ projects by then, understood software engineering deeply, and had built a solid foundation-it was a game-changer. Instead of rejecting or over-relying on it, I focused on exploring its strengths and weaknesses, learning the tricks, and figuring out where it excels and where it falters. Now, two years into my career, I barely write any code myself. Instead, I: Architect the project: I break it down into modules, ensuring the design aligns with my vision. Delegate coding to ChatGPT: I use system prompts tailored to enforce specific coding standards-commenting style, logging practices, level of abstraction, adherence to DRY principles, and testing guidelines. Review the output: Most of the code is up to par, so my role is to ensure it integrates seamlessly with other modules and aligns with the big picture. This approach allows me to build projects that used to take 1-2 days in just 5-6 hours. The key? Knowing exactly how to use LLMs as tools, not as shortcuts. A Warning to Senior Engineers To senior engineers who dismissed LLMs as "useless": those of us who know how to leverage these tools will replace you. Why? Because the game has changed. Efficiency and adaptability matter as much as expertise, and LLMs amplify both if you know what you're doing. A Warning to Novice Programmers To beginners overly reliant on LLMs: don’t make the mistake of skipping the fundamentals. LLMs are not capable of true human-level thinking. They lack critical abilities like complex problem-solving, big-picture-to-small-detail reasoning, and pattern recognition. Unless you master computational problem-solving and the principles of software engineering, you’ll hit a growth barrier sooner than you expect-and it’ll be a steep one. Why People Like Me Are Thriving I (and many like me) have struck a balance: we’ve combined a solid understanding of software engineering fundamentals with the ability to leverage LLMs as accelerators. We know what these tools can and cannot do, and we use them to focus on higher-order tasks rather than grunt work. The takeaway? Whether you’re a senior engineer or a novice, adapt to this new reality-or risk being left behind.

"Driving with a broken headlight,Not fun"

Ai can be helpful as a "quicker google" Dont go with "GPT, I need x or y feature with these parameters write it for me" Instead go with "What are some best pratices or ways to do x" Youll get at least one solution. Then you can go from there to build your own.

I struggeld to start learning programing for years, it feeled like i am to far away from it. As soon as i heard you can AI code your projects, i tryed it. But it never worked, the way i wanted. So i had to read the code, google it, let AI explain it to me, until i could fully understand modify and have a success. I am having a really hard time to code with out AI. But i would have never started to code without AI.

I remember the first time I got a value from a database to show on a screen in a .Net web forms app. After HOURS of trying, it was such a good feeling. Even if the code was questionable. Just messing around till it worked 🎉

but 90% of the time its just fixing AI's code itself cause half of it just garbage code

Golden content, thanks a lot👍

great video, thank you! subbed :) 💜

I often write my code and if I don't have idea after some time what is wrong I ask ChatGPT. I never ask AI to write code and never if I can visualize variables. I asked ChatGPT for my AI model for recognizing digits. And I used it when I tried to make global variable in c++ named max. And I had using namespace std; I don't remember anything else For debugging. I ask for syntax often

I listened to this while going to work, I got my first scripting task, I fucked around till I got it with a little bit of research I usually try too many times to solve the problem but lately I was getting lazy and took easy routes using the GPT but listening to this video gave me a kick to get back to the old days Thank you

underrated video!

F around and find out.

Me watching this to build an todo app

I disagree with this video

Watching this before I go pick up my first set of business cards. In Omnia Paratus.

No thanks. Any issues were easy to fix. Using AI to code I've picked up coding again after my intro courses in college 10 years ago. Ive learned how to properly interpret C++ and Python code. For the most part anyway. I've learned general structure and syntax. Ive learned how to properly write unit tests and take into account edge cases. Ive made afulll RAG pipeline that is amazing for my project management's RL job. Ive forked an Arduino library to include HRTIM functionality for the Giga R1. I've worked on reverse hardware engineering extensively using this method as well.

No clue what yall talking about tbf the code AI writes is pretty good. I only rarely get hallucinated code from Ai even then its because of my shitty promts. Sometimes I even learn new optimisation or some obscure functions I never knew to save me a few lines.

I used AI to convert my cli python project to a GO project. Never used GO before. But not the entire project. I aks to convert only some functions and methods. And aks methods to organize a GO project. Of course hallucinations appeared, but was easy to fix it all, because i knew the behavior i wanted from the project.

Ladies lets fuck until it works 😅

I started some free online course in js, and after first lesson girl asked could gpt write me that..........

I'm a professional copywriter and an amateur/hobbyist programmer. Every single thing you said applies equally to words and code. Great job saying so many important truths and giving concrete examples people can actually follow. Kudos. I'm writing this comment to support you spiritually and algorithmically.

AI sucks

yess my main problem is that i struggle with how i think(logically speaking) when programming 😭 i am also dyslexic so it takes a lot of brainpower and time to do what other people would consider simple 🥹

Asking AI to write you code makes your brain go Jelly, you forget to be able to do it on your own and when you get to the interview stage in person for a new job, you are royally fucked. AI has its uses, don't become some jelly spoon feed programmer, we should always learn problem solving and be able to fix any problem on our own.

When I fucked around with MS Dos I used to Google for parts of the code that I needed and ended in alot of trial and error. Funny enough, one project I couldn't solve with googling or trying other code, actually got solved by just fucking around, trying things that should work in my head. And sometimes, like with if statements, the easiest solution is the one that works.

I think asking AI is only good once you know what your doing and can understand what its doing. But i have leaned all my language by starting with working examples and working on top of them so slowly learn what its doing, so ai is a good start to get that initial starting point. Just depends on how you learn to code. The standard ways of learning have never worked for me cause i get instantly bored with the basics, give me the working code and let me fuck around and find out.