• Post Reply Bookmark Topic Watch Topic
  • New Topic
programming forums Java Mobile Certification Databases Caching Books Engineering Micro Controllers OS Languages Paradigms IDEs Build Tools Frameworks Application Servers Open Source This Site Careers Other Pie Elite all forums
this forum made possible by our volunteer staff, including ...
Marshals:
  • Campbell Ritchie
  • Jeanne Boyarsky
  • Ron McLeod
  • Paul Clapham
  • Liutauras Vilda
Sheriffs:
  • paul wheaton
  • Rob Spoor
  • Devaka Cooray
Saloon Keepers:
  • Stephan van Hulst
  • Tim Holloway
  • Carey Brown
  • Frits Walraven
  • Tim Moores
Bartenders:
  • Mikalai Zaikin

AI tools or websites

 
Ranch Hand
Posts: 34
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Can someone help me with list of AI tools or websites used for programming, debugging, checking sql optimization or development ???
 
Saloon Keeper
Posts: 10779
86
Eclipse IDE Firefox Browser MySQL Database VI Editor Java Windows ChatGPT
  • Likes 1
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
I've been using OpenAI's ChatGPT for Java development for a while now. I'm using the free version but they also have a subscription available.

I've found I've got to work with GPT, that  GPT by itself is insufficient. Debugging is typically your problem but once you point out the problematic code to GPT then it is usually able  to patch the code in short order. GPT doesn't have access to a compiler so can't even get compilation error messages unless you provide them.

I was working on some code with GPT the other night and in an hour I was able to put together an app that probably would have taken me 2+ hours to complete by my self. This was code that I've created bits and pieces of in various projects over the years but I would have had to pause numerous times to reference Java APIs and occasional examples whereas GPT was able to throw the framework together in less than 30 seconds. The initial cut made use of anonymous classes and at some point I asked GPT to use lambdas in all places appropriate and it swept through all the code making the replacements. Unfortunately one lambda expression caused a compiler error which I passed on to GPT and GPT made the necessary corrections. There was one place where GPT couldn't come up  with a formula for the size of one of the JPanels and I had to modify the code myself and give the code snippet back to GPT that  added it to the app. Then there were the usual feature requests and refinements I asked GPT to add, and as long as my instructions were clear enough, GPT succeeded nicely.

So, not a perfect system, but I highly recommend working with AI and becoming familiar with it because that's the future, and you may find you learn things and save time.

P.S. I believe they also have a plug-in for Eclipse but it requires the GPT subscription.
 
Bartender
Posts: 1359
39
IBM DB2 Netbeans IDE Spring Java
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Thanks Carey for sharing your experience. Some colleagues of mine are quite afraid that AI in general will replace programmers.
For me, it isn't a fear I feel. My point of view is that at the current state of the art (and, reasonably, for a very long time, until a different approach will be found to create and train a real general AI), tools like chat GPT and alike will follow the same destiny which was followed by calculators: calculators let humas to rid of calculations made by hand, AI tools will avoid human programmers to write by hand tons of code.
I think that your way of using AI as a code helper will be a common work scenario in the days to come.
What do you think about ?
 
Saloon Keeper
Posts: 27851
196
Android Eclipse IDE Tomcat Server Redhat Java Linux
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
I'm of a like mind, and Carey's observations are most informative to me. But this has nothing to do with JDBC and databases, so I'm going to move this thread to our engineering/AI forum.
 
Carey Brown
Saloon Keeper
Posts: 10779
86
Eclipse IDE Firefox Browser MySQL Database VI Editor Java Windows ChatGPT
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Someone posted this on the web:

I’ve described it as paired programming with a junior developer who has memorized every bit of documentation and who can answer questions about it and who can write pseudo code quickly but who needs someone who understands what it all means to put it together.


Seeing as how my memory is not as good as it once was this fills a gap for me.
 
Carey Brown
Saloon Keeper
Posts: 10779
86
Eclipse IDE Firefox Browser MySQL Database VI Editor Java Windows ChatGPT
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
I haven't figured out how to work with GPT on a project running several thousands of lines long yet. Probably need the subscription and probably the Eclipse plugin.  May also need to feed GPT logical chunks to work on.
 
Carey Brown
Saloon Keeper
Posts: 10779
86
Eclipse IDE Firefox Browser MySQL Database VI Editor Java Windows ChatGPT
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
From GPT:
[Y]ou cannot simply cut and paste a 2000-line Java project into a GPT-3 prompt. GPT-3 has limitations in terms of the input length it can handle, and it is primarily designed for processing natural language text rather than entire codebases.

GPT-3's maximum token limit for a single request is typically around 4096 tokens, and the model's tokenization may not work well with complex programming languages like Java.

If you want assistance with a specific problem or task related to your Java project, you can provide a concise description or a specific code snippet as input to GPT-3, and it may provide suggestions or explanations based on that input. However, attempting to process an entire Java project with thousands of lines of code would likely not yield meaningful results within the token limit and may not be the most efficient approach.

For assistance with a large Java codebase, it's usually better to work with traditional development and debugging tools, collaborate with experienced developers, or consider code analysis and refactoring tools that are specifically designed to handle large codebases.
 
Bartender
Posts: 1156
20
Mac OS X IntelliJ IDE Oracle Spring VI Editor Tomcat Server Redhat Java Linux
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Sadly I've not got the time but these links were posted in Linked-in:

1. AI Essentials - https://explore.skillbuilder.aws/learn/course/external/view/elearning/17763/foundations-of-prompt-engineering
2. ChatGPT Mastery - https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/
3. Google AI Magic - https://www.cloudskillsboost.google/course_templates/536
4. Harvard AI Introduction - https://pll.harvard.edu/course/cs50s-introduction-artificial-intelligence-python/2023-05
5. Microsoft AI Basics - https://www.linkedin.com/learning/what-is-generative-ai/generative-ai-is-a-tool-in-service-of-humanity
6. Prompt Engineering Pro - https://learnprompting.org
7. Google's Ethical AI - https://www.cloudskillsboost.google/course_templates/554
8. Machine Learning by Harvard - https://pll.harvard.edu/course/data-science-machine-learning
9. Language Models by LangChain - https://www.deeplearning.ai/short-courses/langchain-for-llm-application-development/
10. Bing Chat Applications - https://www.linkedin.com/learning/streamlining-your-work-with-microsoft-bing-chat/put-your-fingers-to-work-chatting-as-a-productivity-tool
11. Generative AI by Microsoft - https://learn.microsoft.com/en-us/training/paths/introduction-generative-ai/
12. Amazon's AI Strategy - https://explore.skillbuilder.aws/learn/public/learning_plan/view/1909/generative-ai-learning-plan-for-decision-makers
13. AI for Everyone - https://www.deeplearning.ai/courses/generative-ai-for-everyone/
14. AWS AI Foundations - https://www.coursera.org/learn/generative-ai-with-llms
15. Spring AI - https://www.youtube.com/watch?v=aNKDoiOUo9M
16 - Encode Club - https://www.encode.club/ai-foundation
 
Bartender
Posts: 279
12
IntelliJ IDE Linux
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Sometimes I'll use www.perplexity.ai for quick and simple stuff. I'm kind of surprised nobody seems to know about that one, I never see it in recommendations.

It's nicer because it doesn't force you to get an annoying account... you just use it.
 
Lou Hamers
Bartender
Posts: 279
12
IntelliJ IDE Linux
  • Likes 1
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator

Claude Moore wrote:Some colleagues of mine are quite afraid that AI in general will replace programmers.
For me, it isn't a fear I feel.



I agree with this myself. I've heard some manager types say they expect to be able to fire most developers in a few years and just pay someone to copy-paste from a GPT box. Yeah, right--I don't see that at all.

These so-called AI are okay at cranking out simple isolated code blocks fast, yes. But that's not even close to everything developers need to do. In fact that code monkey stuff is the EASY part of development. To "replace developers" they'll need to do LOTS more.

Let's see them take a large pre-existing project with one or two dozen subprojects, modify it to implement requirements from customers who don't even fully understand what they want the software to do, modify dependencies or update the build, and deploy that thing to a cloud environment with zero downtime without bankrupting the "employer". Oh, and also figure out what to do when random environmental stuff like the network breaks down temporarily (something which never happens in IT, right?). Heck with a lot of these old projects, even getting the build to work correctly can be a challenge sometimes. Good luck, bots!

Maybe some day this stuff will become irrelevant and these ML tools will be able to do something that negates the need for most developers (not all), but I'm not convinced it's coming before I've retired.
 
Consider Paul's rocket mass heater.
reply
    Bookmark Topic Watch Topic
  • New Topic