ChatGPT can create awesome EA? - page 4

New comment
 
Amir Yacoby #:

15 months later, I think it's time to re-iterate and make a statement because too many forum members were treating AI as some kind of "Don't post AI code" or being cynical about it.

So the statement is: The programming profession is dead. It started in the forties of the 20th century, and today, towards the end of the 3rd decade in the 21st century - it's dead after 80 years. 

No human being can cope with this monster. Coding on the highest possible level which two years ago took 50 programmers to complete in a year, can be done by one person in the course of few weeks.

Dead in the meaning that two years ago the best human programmer can not compete with todays models, not even close. That's a mega revolution. 

For more than one year now, there's no need to open the IDE at all - not needing to write or review a single line of code - while being able to write 100X faster, better, near perfection code.

Everybody is free to keep on with manual coding, like you can go with a horse and carriage on a high way. 

Each new generation / model released on those 15 months was jaw dropping and it doesn't seem to stop.

The main thing is that everyone's limits of personal coding have been pushed so far that there's almost no words left. 

 

I think it's exaggerated. The profession is revolutionized for sure, but dead certainly not. Well maybe it depends what is intended exactly with "programming profession".

Anyway, I was skeptical last year or even some months ago and I was wrong. The evolution of the latest months and the progress, specifically with MQL5, are spectacular and it surprised me.

Now, with an updated mindset, professional programmers can benefit greatly of AI co-workers and allow to build big projects a lot more easily and quickly. It's amazing.

 

Where will AI users without programming experience get their experience and qualifications? To be proficient in programming, you need programming experience. Code generation doesn't provide that experience.

I've heard that if an experienced programmer takes a year off, it's hard for them to return to programming. Does this mean that experienced programmers who have become AI users (and stopped coding by hand) will degrade?

 
Vladislav Boyko #:

Where will AI users without programming experience get their experience and qualifications? To be proficient in programming, you need programming experience. Code generation doesn't provide that experience.

Except AI users don't need to program. They ask what they need, check the results and ask for modifications when needed, it's an iterative process.

I've heard that if an experienced programmer takes a year off, it's hard for them to return to programming. Does this mean that experienced programmers who have become AI users (and stopped coding by hand) will degrade?

Never heard or feel that, it's really strange to me. 

Anyway, experienced programmers, from my experience, doesn't stop coding by hand, they do use AI and also code "by hand". It's not one or the other. You can delegate the more trivial, boring or repetitive tasks to the AI, then focus on the highest level tasks. Well, I am coding alone, so within a team it could be different, don't know.

 
Alain Verleyen #:
Never heard or feel that, it's really strange to me.

I heard this from an employee at an outsourcing IT company. Perhaps they were working at a high intensity, I don't know.

Alain Verleyen #:
Anyway, experienced programmers, from my experience, doesn't stop coding by hand, they do use AI and also code "by hand". It's not one or the other. You can delegate the more trivial, boring or repetitive tasks to the AI, then focus on the highest level tasks. Well, I am coding alone, so within a team it could be different, don't know.

I completely agree with this.

I meant the extreme when a person completely stops programming manually. This is probably what is meant by those who say that AI will replace programmers or that programmers will start writing prompts instead of code.

 
Alain Verleyen #:

I think it's exaggerated. The profession is revolutionized for sure, but dead certainly not. Well maybe it depends what is intended exactly with "programming profession".

Anyway, I was skeptical last year or even some months ago and I was wrong. The evolution of the latest months and the progress, specifically with MQL5, are spectacular and it surprised me.

Now, with an updated mindset, professional programmers can benefit greatly of AI co-workers and allow to build big projects a lot more easily and quickly. It's amazing.

It is dead in the sense that a machine can do the job better than any programmer, meaning that a programmer have much less value, if any - but its decreasing fast, to an employer.

And also, it's job is no more in writing code, fixing bugs, understanding code etc. 

Most hi tech companies are already shifted to fully (no progrmmer writes code) or are in the process of doing that. And in any case, the programmers themselves use AI to write the code, offically or not.

 
Vladislav Boyko #:

Where will AI users without programming experience get their experience and qualifications? To be proficient in programming, you need programming experience. Code generation doesn't provide that experience.

I've heard that if an experienced programmer takes a year off, it's hard for them to return to programming. Does this mean that experienced programmers who have become AI users (and stopped coding by hand) will degrade?

I can approve, after one and a half years of not writing at all (mql5) but getting much more than before, when I try to code at work (still doing it in my company, though they are fastly progressing towards ending it) - I struggle a lot, it feels stupid for me to do that because I realized the futility in doing it.
 
Alain Verleyen #:
Except AI users don't need to program. They ask what they need, check the results and ask for modifications when needed, it's an iterative process.

Never heard or feel that, it's really strange to me. 

Anyway, experienced programmers, from my experience, doesn't stop coding by hand, they do use AI and also code "by hand". It's not one or the other. You can delegate the more trivial, boring or repetitive tasks to the AI, then focus on the highest level tasks. Well, I am coding alone, so within a team it could be different, don't know.

I think you still don't get it right.
It's not the boring routine tasks. Say for instance - write me a visual editor in mql5? Is it a boring routine task? Surely not. Are most people able to do it in a year - I doubt it. 
So? You can do it very fast today.

If I tell you it can be done in few weeks, without writing a single line of code, and in parallel do additional big tasks - does that mean the programming proffession is dead? 

In my eyes - certainly.
 
The thing is, that today everyone can build large things, but investors are harder to impress - because anyone knows that it can be done again with AI - so why pay? Or better said, the 1 MIL USD paid by investors to a product before the AI era, will maybe yield a 100K USD today. 

The software is becoming cheap. Libraries are easy replaceable and don't deserve so much human attention. 

That's my view.
 

In my opinion, there's a big mistake: trading has never been about programming. Creating an expert advisor was difficult for those who didn't know how to program, but freelancers have always existed.
In trading, the problem is FINDING AN EDGE.
AI can't find that for you. 
You can spend all day making it program and following its advice, but they will be losers.
Just like with SAS... The problem was never creating the software, but marketing it.

So can CHATGPT create AWESOME EA? No, people create AWESOME EA with chatgpt if they already knew what they were doing before. Otherwise, it just creates garbage, but faster.

 
Liang Zhang:
I am very interested in whether ChatGPT can create a very intelligent EA. Has anyone used ChatGPT to write an EA?
In my experience AI is a useful tool,  but unfortunately it isn't a substitute for understanding trading logic. I used AI to help with parts of development, but every trading rule, risk management decision, debugging session, and validation step still required manual design and testing. The quality of an EA ultimately depends on the trading strategy, testing methodology, and developer insight.
12345678
New comment