Show HN: AI-powered code correction that teaches you along the way (opens in new tab)

(useadrenaline.com)

81 pointsjshobrook3y ago22 comments

22 comments

I fired up the browser DevTools to see how this works. It submits the code directly to the OpenAI API (I like that it does this in the browser rather than forwarding my API key on to the useadrenaline.com server where it might end up logged) with the following options:

    {
        "model": "code-davinci-edit-001",
        "input": "### Code goes here ###",
        "instruction": "Identify and fix all bugs in this Python code."
    }

Then the application itself has a nice implementation of client-side diff presentation.

That was for the "lint" button - I didn't run this experiment for the "debug" button.

hgsgm3y ago

> model": "code-davinci-edit-001"

Is that Adrenaline's own model (OpenAI let's you train your own model in their environment, right?), or OpenAI's model?

frabjoused3y ago

The logo is directly from Adobe's book: https://www.adobe.com/products/catalog.html#category=creativ....

I would recommend changing it.

ParadisoShlee3y ago

I tested this one some ruby, and it told me to break my code with typos. -- Replace "enable_starttls_auto" with "enable_starttls_aauto".

I will be super interested when I can run this against a whole git repo codebase instead of a single file.

ilaksh3y ago

I have an experiment related to that which seemed to mostly work. What I did was give it the project dir listing first along with the request and ask which files it needed. The other part of it was to give it a specific format to list the file updates with their file name first.

What I will probably actually do when I get a chance for aidev.codes is use OpenAIs embeddings with a vector search for relevant snippets for the prompt context. Or possibly use the gpt-index project which I think does that for me.

aogaili3y ago

Help me to understand but why would I use this as opposed to just pasting the code to ChatGPT?

Also, it would be nice if there is a WebStorm plugin.

woah3y ago

This is why it may be tough for startups without in-house ML expertise, proprietary weights, and the resources to continually train them to retain a moat. Innovations on the base layer seem to eat light packaging layers for breakfast. I knew a guy who got early access to the GPT-3 beta a few years ago and made a site where you could upload apartment leases and get an explanation. Now with chatGPT you can just ask it for what you want it to do. Of course, both this guy's GPT-3 wrapper app, and chatGPT both give inaccurate but dangerously plausible sounding answers, but that's a different problem.

Ozzie_osman3y ago

Because good products meet you where you are, in your workflow, as seamlessly as possible.

Also, presumably they will continue to refine beyond what ChatGPT is doing with better models, smarter prompts, etc.

kanyethegreat3y ago

This is literally a web app. It’s not meeting me anywhere OpenAI’s web app isn’t. There are VS code plugins that people have built that put ChatGPT in your IDE by reverse engineering the API (I wrote one).

Also, without access to the model (eg. by merely calling OpenAI’s API), no one’s refining it. They need access to the actual model (eg. Bloom)

throwaway6753093y ago

Not with their current implementation, it's literally just making API calls to OpenAI GPT models. They can improve their prompts, but it'll never be better than what OpenAI offers as a first party.

412093y ago

I would love this as a Visual Studio or VS Code plugin.

Using it as a standalone website looks fun too, can't wait to dig into this !

kanyethegreat3y ago

Check the marketplace. They exist.

jackson14423y ago

Doesn't seem to work after pasting API key (just keeps asking for API key over and over). Also seems to force email input to subscribe to a mailchimp list.

You should also set your site to force HTTPS.

ilaksh3y ago

Has anyone been able to get OpenAI to increase their rate limit for Codex (code-davinci-002)?

I have a somewhat related service https://aidev.codes but I have to default to using text-davinci-003 instead because the code-davinci-002 rate limit is very small (10-20 requests per minute).

I have been trying to contact their support about it for a month without any response.

crummy3y ago

I tried this but after entering my API key then clicking Debug, it just loads forever. Console says:

xhr.js:162 Refused to set unsafe header "User-Agent"

POST https://api.openai.com/v1/edits 429

jshobrookOP3y ago

Surprisingly helpful with custom exceptions, nice job

mdrabla3y ago

Thank you! We think that documentation in - code out would be a cool feature as well.

ipgmonstereater3y ago

This is so useful, I just need ai to help me through my design too lol

lumost3y ago

Give your problem statement requirements and context and see what chatGPT thinks bout your design. It’s halfway decent from what I’ve seen on open source projects.

Obertr3y ago

which kind of problem are you facing?

leaving3y ago

This seems nice and quite simple, but I think I'll wait until there is an alternative to OpenAI that doesn't censor what I can do and impose the morality of a profit-making corporation on my work.

Considering that the morality of capitalism is indistinguishable from the morality of a cancer cell, I think I might be on the right side of history here.

I'm also not keen on giving up more of my personal information to use this.

ramen_rams3y ago

Nice job!

j / k navigate · click thread line to collapse

22 comments

simonw3y ago

    {
        "model": "code-davinci-edit-001",
        "input": "### Code goes here ###",
        "instruction": "Identify and fix all bugs in this Python code."
    }

Then the application itself has a nice implementation of client-side diff presentation.

That was for the "lint" button - I didn't run this experiment for the "debug" button.

hgsgm3y ago

> model": "code-davinci-edit-001"

Is that Adrenaline's own model (OpenAI let's you train your own model in their environment, right?), or OpenAI's model?

frabjoused3y ago

The logo is directly from Adobe's book: https://www.adobe.com/products/catalog.html#category=creativ....

I would recommend changing it.

ParadisoShlee3y ago

I tested this one some ruby, and it told me to break my code with typos. -- Replace "enable_starttls_auto" with "enable_starttls_aauto".

I will be super interested when I can run this against a whole git repo codebase instead of a single file.

ilaksh3y ago

aogaili3y ago

Help me to understand but why would I use this as opposed to just pasting the code to ChatGPT?

Also, it would be nice if there is a WebStorm plugin.

woah3y ago

Ozzie_osman3y ago

Because good products meet you where you are, in your workflow, as seamlessly as possible.

Also, presumably they will continue to refine beyond what ChatGPT is doing with better models, smarter prompts, etc.

kanyethegreat3y ago

Also, without access to the model (eg. by merely calling OpenAI’s API), no one’s refining it. They need access to the actual model (eg. Bloom)

throwaway6753093y ago

Not with their current implementation, it's literally just making API calls to OpenAI GPT models. They can improve their prompts, but it'll never be better than what OpenAI offers as a first party.

412093y ago

I would love this as a Visual Studio or VS Code plugin.

Using it as a standalone website looks fun too, can't wait to dig into this !

kanyethegreat3y ago

Check the marketplace. They exist.

jackson14423y ago

Doesn't seem to work after pasting API key (just keeps asking for API key over and over). Also seems to force email input to subscribe to a mailchimp list.

You should also set your site to force HTTPS.

ilaksh3y ago

Has anyone been able to get OpenAI to increase their rate limit for Codex (code-davinci-002)?

I have a somewhat related service https://aidev.codes but I have to default to using text-davinci-003 instead because the code-davinci-002 rate limit is very small (10-20 requests per minute).

I have been trying to contact their support about it for a month without any response.

crummy3y ago

I tried this but after entering my API key then clicking Debug, it just loads forever. Console says:

xhr.js:162 Refused to set unsafe header "User-Agent"

POST https://api.openai.com/v1/edits 429

jshobrookOP3y ago

Surprisingly helpful with custom exceptions, nice job

mdrabla3y ago

Thank you! We think that documentation in - code out would be a cool feature as well.

ipgmonstereater3y ago

This is so useful, I just need ai to help me through my design too lol

lumost3y ago

Give your problem statement requirements and context and see what chatGPT thinks bout your design. It’s halfway decent from what I’ve seen on open source projects.

Obertr3y ago

which kind of problem are you facing?

leaving3y ago

This seems nice and quite simple, but I think I'll wait until there is an alternative to OpenAI that doesn't censor what I can do and impose the morality of a profit-making corporation on my work.

Considering that the morality of capitalism is indistinguishable from the morality of a cancer cell, I think I might be on the right side of history here.

I'm also not keen on giving up more of my personal information to use this.

ramen_rams3y ago

Nice job!

j / k navigate · click thread line to collapse