My current one is https://limnology.co/ , organizing the top 70k educational youtubers in 20 languages. Youtube's channel discovery is pretty bad, and I wanted an easy way to find new channels about the topics I'm interested in. Still a work in progress as the problem turned out to be a lot harder than I thought.
Unfortunately all the other languages have too few results, I just checked and Greek only has 0.7% as many channels as English in the db, so it'd only be a few pages of content
I did think about it, but my concern was that because I only take a small subset of top keywords for each channel, it's very easy to create a combination of tags that'll show no results. I'll play around with it and see if it makes sense to implement.
Because it's a "list"? I don't think that's correct, there's ballpark 200k different pages with unique content in that app, that's a lot more than just a list.
I had to split it because the sheer size of Indian audiences breaks all the metrics, and yet those videos would not be super relevant in the west as a very large percentage of the Indian educational youtube consists of various kinds of local exam/test preparation channels
Try digging deeper into the "Related Keywords" cloud at the top, the narrower is the keyword you pick, the more long-form & technical and the less pop-channell-y stuff you'll see.
I'll add a report button, it definitely needs it. All of that data is from the the youtube channels themselves and _a lot_ of them mislabel things, even such basic ones as their channel's language.
Yeah, that is something I'd love to implement but so far could not figure it out, even simply interlinking the keywords across 20 different languages is very non-trivial
Elon is one of the best people to learn from about the development speed, given how fast they move at SpaceX, so I thought I'd post it here.
My summary of what he talks about in his 5 steps:
1) Make the requirements less dumb. I.e. always question the requirements and always have a name of a person responsible attached to them (not the department), so you know whom to question.
2) Try very hard to delete a part or a process. If you are not adding things back 10% of the time, you are not deleting enough.
3) Simplify and optimize, but ONLY after doing the first 2 steps.
4) Accelerate cycle time, you can always go faster.
5) Automate.
All five only in that order, he talks about making the mistake of going through the steps backwards and then deleting the part.
Overall it's a great interview at the Starbase Factory, so I recommend you watch the whole thing, or at the very least the relevant ~12min portion.
I was curious about freedom of travel benefits and made an app where you can check how many more countries you'd be able to access with another passport on top of your existing one: https://multinational.io/best_passport
Build an app for finding lost pets with facial recognition. Recognizing cats is the easiest thing for neural nets, yet I haven't seen it used for lost pets yet.
Basic idea, if my pet is lost, I open up your app, upload a few images of my cat and set a monetary reward. On the backend your neural net learns how that specific cat looks like, then other users of the app can snap pictures of random cats on the street and the app will tell them if that animal is lost or not. If someone snaps a picure of a lost cat, they'll get connected to the owner and the reward is transfered, the app can take a small cut of it.
Does anyone have any experience with soft robotics? For example these guys: https://www.youtube.com/watch?v=X6CRe2ieuYE advertise their gripper as supposedly being able to handle weight/size variety with no training at all, just with the use of different materials in the gripper