AI keto coach

@dmytro_konovalov1 Congrats on your launch! I am excited to try out your product. I received a privacy error warning while visiting your site (via the link on this post). Seems like a certificate issue. Hope you get that fixed soon

@goyashy Wow nice to see you :) Your photo is perfect :) I'm curious, how does the chat-format diet onboarding work exactly? Imagine you are starting keto diet, and you only know only basic idea - cut carbs and count calories. You have read some articles (maybe 2 or 3) on the internet maybe some on reddit, so right now you feel a bit more educated and striving for making you first week carb free. So here what happened: during this first exploration you didn't even knew what to explore, so you learned bare minimum without real advices. So you getting keto flue (extreme fatigue and headache) which makes you seek and removes all your willpower to continue, so you are giving up within first 3 days. I know it personally and I know a lot folks who have hard time on keto. The problem was simple - lack of a lot of useful information about keto basic at the very begging that will just make your life sooo easier. And the root of the problem is simple to - internet in nowadays format (youtube, reddit, even ticktock) is unfortunately not the best and personalized format to explore actually any topic. I can uncover this topic and have wider discussion 🤗 And the only solution I have in my head right now is make information tailored for specific needs with the power of GPTs. Friendly simple motivational, and give you right opportunities to explore and be curious. And can you share some of the interesting ideas you have in mind to make it even smarter? 😊 Sure the format I have right now is particulary dependent on general knowledge of GPT and "statements" about keto diet from internet articles that are validated during truth checking question-answer algorythm (here some know how: I made specific GPTs that are asking questions related to information and answer them determining if information is optimally true). Right now "statements" db is kinda small for the topic - 300 statements, actual smartness however can be easily done with incorporating human feedback and expading db to 1000x scale :) Hope my answer was useful :) Maybe you want to know something else? :)