Live data from Hacker News

Show HN: I've been using AI to analyze every supplement on the market

pillser.com

31–40 of 50 posts

Re: Show HN: I've been using AI to analyze every supplement on the market

#31

I've fucked myself up so much over the years with supplements. Wish I could go back 30 years and tell myself to just eat real food. B6 toxicity for instance is crippling. That being said these days I just take two things: vit K2 and magnesium.

That's solid advice for most people in the Western world. K2 and magnesium are sensible choices–genuinely hard to get enough of through diet alone. The bigger problem is that most supplement sales are marketing-led. That's what pushed me to at least try to make the whole process more scientific.

Re: Show HN: I've been using AI to analyze every supplement on the market

#32

Please don't take any products down due to shady legal threats! Won't this compromise nearly the entirety of the value proposition? Cripes.

I don't want to, and I don't plan to. At a time when this happened, I was deep in my studies preparing for exams, and I just remember thinking to myself, "I cannot afford to pull myself into what could become a legal matter" I am now in better position to allocate time and attention should such claims be made again. That said, the couple of brands that were removed were not brands that I would have wanted anyone to b…

Maybe, have a page in which you show that all the details of that product have been redacted as the producer requested. Something like a DMCA strike list. This leaves a great say about how muche the company wants to manage the narrative of their product.

Re: Show HN: I've been using AI to analyze every supplement on the market

#33

  U GOT A WAYK00L BB0ARD D00DZ!

  I borowed my brothers vic-2O to read ur bboard!!1!

  Suplements r important 2 me after mom fed us nothing but enfamil

  It looks like my doc says i got "Irritable Biff Syndrome"

  So I gonna order some of these (do they accept Minecraft or RuneScape money??

  UGOT WAYK00L BB0ARDZ, D00DZ!!
B1FF B1FF

   B1FF B1FF

       B1FF ... B1FF
!!! B1FF ‽ B1FF

Re: Show HN: I've been using AI to analyze every supplement on the market

#35
I just searched the most popular/researched supplement of all time: creatine. There is a mistake in the data there: the Wellnesss Code Whey Protein indeed contains creatine, but not 2g per container, but 2g per serving (that is correctly reflected). Error is easily spotted due to the price per gram being an extreme outlier. That is perhaps something you can look for when evaluating the data gotten from the LLM.

Re: Show HN: I've been using AI to analyze every supplement on the market

#37
Please seriously consider *not* using ads to generate revenue. We need to return to other models rather than just slapping ads on everything. Be the change you want to see in the world. Do you really want to see more ads?

Have a look at https://examine.com as the closest site I can think of in this space, and check out their revenue models.

Re: Show HN: I've been using AI to analyze every supplement on the market

#38
post #4

This seems more like cataloguing than analyzing I'd love to see a project that actually analyzes every supplement on the market to make sure it actually contains what it claims to, contains it at the listed dosage, and to show anything else found (heavy metals for example). That's not something AI can do for us though since it'd involve physically collecting and testing samples.

That's my dream! That's where I want to take this project. At the moment, my focus is on what I can do by aggregating and analyzing the available data. Extracting ingredients, normalizing them, normalizing quantities across every supplement, etc. This has already proven to be a lot bigger undertaking than I could have imagined at the beginning of this journey. I had learn a lot about databases, scraping, LLMs, and ev…

I did something like this, pre LLM, except with food additives.While I dont have the technical expertise to fully evaulate what these substances are, and do, in many cases they are banned in some juristictions, and often have exceptionaly dubious reasons for bieng put in food, I did notice one very stark thing, in that not one single food aditive has ever been held up as " WOW! this stuff is so good!, a boon and benifit, etc, whatever" nothing, zero substances acclaimed as unequivical benifit to humanity. I eat only food now. Oh, and one more thing, many of these things are "white crystaline" substances, which means they are 100% concentrated things that are never found in actual foods, think sugar.

Re: Show HN: I've been using AI to analyze every supplement on the market

#39

Please seriously consider *not* using ads to generate revenue. We need to return to other models rather than just slapping ads on everything. Be the change you want to see in the world. Do you really want to see more ads? Have a look at https://examine.com as the closest site I can think of in this space, and check out their revenue models.

I know Examine.com, and I've spoken to the founder a few times. I really appreciate what they are doing.

I am very open about how the site is planning to be monetized https://pillser.com/affiliate-disclosure

Re: Show HN: I've been using AI to analyze every supplement on the market

#40
post #35

I just searched the most popular/researched supplement of all time: creatine. There is a mistake in the data there: the Wellnesss Code Whey Protein indeed contains creatine, but not 2g per container, but 2g per serving (that is correctly reflected). Error is easily spotted due to the price per gram being an extreme outlier. That is perhaps something you can look for when evaluating the data gotten from the LLM.

There is interesting context here. The first version of Pillser was focused on very narrow area of supplements, and specifically, supplements sold in the shape of pills (therefore the name "Pillser" [pill search]). However, it kinda snowballed from there, and powder substances were one of the last things I've added. The evals need to be expanded to have more examples specifically around powder substances. However, that's a fix I know how to implement and will prioritize.

Thank you for sharing the example that you've found!

Post reply on HN