I appreciate you letting me know. I can see it in Sentry as well. It's odd though, it looks like it's coming directly from React. There are no traces in my own codebase. I'll try to isolate it.
That is odd. When I searched for Vitamin D, it showed me a list (which I could sort by $/mg). That list included Vitamin D, Vitamin D2 and Vitamin D3. I assumed (incorrectly it seems) that this meant that it was doing some sort of prefix match. It seems that the result table highlights the "Ingredient" column, so I assumed (incorrectly) that it was searching that column. However, it doesn't seem to search the "Supplement" column, nor is it doing a prefix match (or substring match) on the Ingredient column. This is just confusing.
There is even a Google Sheet that is generated based on the data that we have, although I just noticed that it's not updating properly, so I need to fix that.
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 evals. But it has been a tremendously fun (if sometimes overwhelming) journey.
First, I need to figure out how to monetize what I've built so far. And maybe I cannot, in which case I will start over. But I am trying to find my niche that people uniquely value. So far I found that there are is a pull from people chasing deals (e.g. finding products containing specific ingredients with the highest price per mcg) and people seeking for niche ingredients. This is not exactly the target audience I had in mind when building this, but I am glad they are finding value.
Evolving into performing actual lab tests and producing our own high-quality supplements is my dream.
I would love to get your feedback on what data to prioritize and how to make it easier to browse.
I've learned from my last sprint that this project can get overwhelming. So now I am making it my priority to build a community of people I can reach out to for feedback and direction. If supplements is your thing, I would greatly appreciate being able to chat with you from time to time (even if it just a quick email!)
Hi, not into supplements, but really appreciating your approach to building a community and the reasoning behind it (and that you're being so transparent about it!). I'll give this a go too, thanks for the inspiration :).
My idea is pretty simple – I want to bring scientific proof to the supplement shopping process.
Too many products are sold with exaggerated claims or without any science backing.
I am doing two things to fix it:
I am using LLMs to index all scientific papers that mention different supplements. I then extract insights from those papers in relation to the supplement and summarize on a matrix of effect type relative to the health outcome and effect size, e.g. https://pillser.com/health-outcomes/enhanced-disease-resista....
I then index all ingredients of every supplement on the market, and then based on the composition of the ingredients, I surface the relevant scientific evidence to back or dismiss any health claims, e.g. https://pillser.com/vitamins/vitamin-b1
People are able to search products (https://pillser.com/search) by their desired health outcomes, dietary restrictions, and other attributes important when shopping for supplements.
The business model of Pillser is the affiliate revenue from the outbound links that take people to Amazon, etc.
Pillser has several hundred regular users that use it to ask questions when researching supplements. However, otherwise, the unexpected challenge has been that getting news users is nearly impossible because ranking for related keywords on Google is nearly impossible. So the growth has mostly stalled.
It's been my life's dream to build something in intersection of my interests – nutrition and AI. So I am excited to have built a product that is serving users. However, I am also debating what's the path forward, as the original idea of building a free to use product that starts making money from a large scale of affiliate sales seems to be a foggy dream given the current Google trends.
Proud to share what I have accomplished and would also love your feedback for how to move forward.
Curious to see how it does, as it was something I considered building! The thing that put me off from doing it is that it is just too simple for existing platforms to add AI feature on top of their existing offerings.
Oh interesting! We've definitely seen that its not super easy to just add AI for relevance if you have a super diverse user base. Currently our customers are early stage b2b SaaS and we understand what their goals and use cases are. Therefore, adding something like relevance isn't so straightforward for a big player IMO.
But cool to hear you thought of building something similar! I think its a good sign that there is a need.
It's been 3 months in the making and a journey started with an exchange on HN. I couldn't be happier to share progress!
What is Pillser? Pillser aggregates data from thousands of research papers and about thousands of supplements, bridging the gap between scientific research and consumers. Our goal is to provide easy access to unbiased information on supplements and nutrition, empowering you to make informed health decisions.
Pillser Ask leverages the same data that powers Pillser, but also has access to sources like ods.od.nih.gov, nhs.uk, and a dozen more authoritative sources on nutrition. It can give personalized, evidence-based advice on nutrition and supplementation.
The chat is prompted to work with as little or as much information that you feel comfortable disclosing. However, the more variables that you tell the chat (like age, sex, body weight, etc.), the more of these variables are used to match Pillser's vast database of research papers (https://pillser.com/research-feed) to surface the most relevant insights.
In future, the personalization is going to be built into the product. I am working on a health card – you will be able to opt-in to creating a profile on Pillser that describes your age, sex, body weight, existing medical conditions, etc. and Pillser will automatically take these into consideration when you ask Pillser Ask a question.
A few necessary disclaimers:
Pillser generates revenue through affiliate marketing. We earn a small commission on sales made through our website, but we do not promote any specific product or brand. This ensures we provide unbiased information and insights without the influence of advertising. Full disclosure: https://pillser.com/affiliate-disclosure
Pillser is a young project, just over 3 months in development, and I envision it as my life's work. The field of healthcare, nutrition, & supplements is incredibly complex and nuanced, and I believe AI will play a critical role in making it more accessible. However, in this early stage of Pillser, I am still learning how to best utilize the data that powers it. I am putting a lot of effort into making sure that the information provided is accurate, but mistakes are bound to happen, and I am always open to feedback and suggestions. Learn more: https://pillser.com/ai-powered-supplement-research-and-compa...
I don't know yet how exactly the project will evolve over the time. I just know that there is some intersection between people wanting to eat and live healthier and vast amounts of research, and that somehow successfully combining them can create something that people want.
Would love to hear your perspective on the project today and where your imagination takes you!
reply