aldamiz

Anytime:)

Good job! We built something in the space, discussed it in detail here https://news.ycombinator.com/item?id=21703297

thanks rurp - fixed.

I wouldn't say our "Product people" are very traditional product people, you have our bios and focus in the team link in the post and below. Hope it helps

https://docs.google.com/document/u/1/d/1r0IIViwp5MuNcWXZqNI2...

Yes. I wouldn say our "Product people" are not very traditional product people, you have our bios and focus in the team link in the post and below. Hope that clarifies the doubts. I've also clarified it in the text, and it now reflects the reality better.

https://docs.google.com/document/u/1/d/1r0IIViwp5MuNcWXZqNI2...

thanks sofia :)

thanks theicfire :) maybe in the future!

:)

Hey AndrewKemendo, such a great comment, so many points to think about. Thanks for sharing!

In our case, size is a property in the ontology.

For some players (second hand for example) size is the number one "personalization" requirement. Imagine you see a feed of products, and none of it is your size...

:)

You are right, the fashion taste api website does not intend to be stylish. It tries to convey a message to business and tech teams, and we know it could have a better design. Take a look at the consumer product https://www.youtube.com/watch?v=EMMmdCB1-Wg hopefully you like the design better :)

Thanks bumblebee4 - fashiontasteapi.com is a the b2b side of the company and the value prop is that we can help retailers automatically classify clothes and understand/classify people. It is for technical and business people, not the end consumer. This is a line of business we were just starting, and hopefully will continue.

The consumer website is chicisimo.com :)

Yes you are right, it is an excerpt from one of our patents. I completely agree that it is very confusing as a caption, and I've taken it out. Doing it in a hurry didn't help. Thanks for the heads up.

I hope you can read the post and be more positive. Thanks!

Facebook always linked to great news!

Thanks Andrew. They are an amazing team with a great ontology, I agree. Thanks for the heads up!

Thanks for your interest Goldemerald. A couple of comments:

- We do use DL. I didn't mention it because it is not relevant in the context of the post;

- How we use DL. One of the jobs of the graph is to tell the DL algorithm what content needs what type of descriptors. The graph can do this thanks to the different levels in our ontology, and because it understands our content in its context: it understands people's interaction and tagging. DL is a small part in our entire infrastructure;

- The DL market. Lots of companies use DL to identify attributes in an image, and the level they achieve is impressive. Having had long discussions with the best of these companies, I can tell you that building the correct ontology is nearly impossible without the entire infrastructure. We are happy building the intelligence that tells DL what to do, and then attaches descriptors correctly to outfits and taste profiles;

- Our patents. They cover a few relevant aspects in the online fashion market: a system to tag fashion images with shoppable products; a system and method to capture/understand how people mix and match clothes in outfits and closets; and more.

I wonder how much will this impact in conversion to trials

Hey Andrew, we started building it in 2012. The concept is similar as you say. The most relevant difference is the ontology: our ontology is very specific to what-to-wear needs, theirs isn't. Thanks

>FB will use artificial intelligence to determine when someone has died

Everything about Facebook looks terrible

Totally agree with this "The product mindset is a complicated definition". I'd say it is about understanding that building a product people really love and engages with, is key to success. Not the only factor of course, but that's what many business oriented people do not understand.

If I understand well, you are using both product tags from retailers, and generating your own tags using image search? I'm curious: how many tags have your trained your system for... how large in your taxonomy?

I would totally use this!