dcastm

The title is wrong. There's no effect here, just correlation.

It's pointless to draw any conclusions from this study. If you think otherwise, check https://www.tylervigen.com/spurious-correlations

This works well for simple use cases, but quickly breaks down when faced with real-life scenarios.

Database schemas might contain hundreds of tables, and they may not always have intuitive names. Also, the relationships between the tables aren't always clear, and there are always company-specific mystery clauses that you might need to apply (such as excluding certain users) when running a query.

Anyone building a chat interface for databases has probably realized this by now.

Interesting. But I believe you're be wrong about Show HN (as others have pointed out).

Show HN makes for an effective way to get to the front page, based on three reasons:

1. You don't need that many votes to make it to the front page, and the votes don't need to happen in the first minutes after you've posted.

2. Many people use shownew[1] in addition to newest[2] to discover content on HN.

3. Show HN posts remain visible for a longer time on shownew compared to regular posts on newest.

When you post using Show HN, your post stays for a long time in shownew (right now the last one there was posted 19 hours ago), while on newest your post has to gather votes very quickly to make it to the front page (the last post visible there is from ~50 minutes ago). So Show HN gives you a higher chance of getting your post "discovered".

In my case, I've made it to the front page 4 times out of 21 posts. 3 of those were Show HN posts.

[1] https://news.ycombinator.com/shownew

[2] https://news.ycombinator.com/newest

Tbh I’ve seen AutoGPT mentioned a lot, but haven’t actually seen anyone using it regularly

Not my case but I thought this might help. I wrote some tips on how to use LinkedIn effectively: https://dylancastillo.co/tips-for-standing-out-on-linkedin/

It’s based on published papers about its algorithm + personal experience.

It runs three fairly slow queries on GPT-4, checks if the domains are available, that there aren't any registered trademark issues.

It takes roughly 6 mins to run everything. Maybe a better choice would have been to show the results as they became available.

Oh nice! I'm curious to know how you did the finetuning. I've been thinking about it as well.

You're correct about the approach. Sometimes it mixes the sources for whatever reason, but gotta investigate it more.

For the sourcing, I used a small variation of perplexity.ai's prompt: https://twitter.com/jmilldotdev/status/1600624362394091523/p...

Yes, it uses embeddings (but not from OpenAI) as those are too expensive.

Sorry, wasn't really prepared for that much traffic and got hugged to death for a while :/

Nice!

free-genie is my own API. It also uses gpt, but my goal is to collect data and then finetune a model with it.

What's wrong about it?

It seems to get the result I want, but not sure if I'm missing something.

Haven’t used it. Will check it out, thanks!

Great idea. Providing links to the relevant manpages makes a lot or sense.

It won’t run any command unless you explicitly tell it to do so.

But yes I’d also like to provide some sort of warning if a certain command “looks dangerous”. I’ve been thinking about what's the best approach.

Thanks! I thought of using a shorter name but the ones I liked were already used in pypi.

You can always add `alias genie=shell-genie` to your .bashrc or .zshrc file :)

Cool idea but seems like the kind of product bound to have the same fate as Google glasses or Segway.

Note that the guy who submitted the images to the contest created hundreds of images, and spent weeks finetuning the prompts and curating the results.

Midjourney/Stable Diffusion/DALLE are doing to Photoshop what it did to traditional drawing methods. But there's still a human in the loop.

In the M5 competition[1], most winning solutions used LightGBM. So ML beat classical.

Just a couple of the winning solutions used DL.

[1] https://www.sciencedirect.com/science/article/pii/S016920702...

Looks like she goat herself in trouble

Thank you!

Thank you for all the ideas!

I understand what you mean. Gonna think this more thoroughly. Thanks for the feedback.

You're right, I should make it easier for users to share the game.

I have a very small audience in twitter, and don't tweet that often, so sharing it that way wasn't really making a difference.

This are great ideas. Thank you.

I'm going to implement a few of these.

This is great feedback. I'm going to take this and go back to the drawing board. Thank you.

These are great puzzles!

I used to play Flood a lot so I ended up creating my own version of it with daily challenges: https://fastflood.dylancastillo.co

I tried searching for some exact phrases from the live demo, but didn't get any results in Google.

Maybe Google isn't crawling all the pages on the site? Possibly related to being an SPA?

Otherwise, I think this is a pretty cool project. I've always thought that Slack and Discord communities lack a good way to find them.

If reddit is new digg, what's the new reddit?

Thank you. I've been thinking about that change. It does look cool haha

Thank you!