Nutritionist AI
Free AI-Powered Nutritionist App
Hey Indie Hackers
I’m Gerardo, an entrepreneur from Mexico. I’ve always run physical businesses, but I’ve been quietly obsessed with building a tech product — even though I’m not a developer.
A few months ago, I finally decided to take the leap.
I just launched my first app: Nutritionist AI — a free, AI-powered nutritionist app that generates personalized meal plans in seconds based on user input like age, weight, goals, food preferences, intolerances, and more.
It’s completely free to use — the only thing I charge for (optional) is exporting the results to PDF. You can always view everything inside the app for free.
Why I Built It
It started as a curiosity project. I was playing around with no-code tools and AI-assisted coding, and thought:
“What if I could build something useful — even without knowing how to code?”
My wife is a nutritionist, so originally I was building a tool for her. But then I thought: Why not make it available for anyone who wants free, fast, personalized nutrition help?
That’s how Nutritionist AI was born.
How I Built It (With No Tech Background)
I had never written a line of code before this.
I learned by watching tutorials, using ChatGPT to fix bugs and explain concepts, and hiring a freelance developer to help with the heavy lifting.
The process took over 6 months — way more than I expected — mostly because:
The app needed to handle a lot of input from users (clinical history, food preferences, goals, etc.)
I wanted the results to feel like a real nutrition consultation
App store policies are… intense
But the fact that this app is now live on both App Store and Google Play still blows my mind.
Where I'm at Right Now
Truth? Growth has been tough.
I get less than 10 downloads per week (combined), and almost no feedback yet, which makes it hard to improve.
I think part of the challenge is that I don’t have a strong online audience or community around the app — and that's why I'm here on Indie Hackers.
How It Works (In 10 Seconds)
User enters personal info: age, sex, weight, height, allergies, food likes/dislikes, goals
The app generates 3 meal options per mealtime (e.g., breakfast, lunch, dinner)
Users can view their plan instantly (no charge), or optionally pay to export it as a PDF
Hardest Part?
Getting everything to actually work
Even with help, AI tools, and tutorials, it was harder than expected to:
Handle all user input properly
Generate consistent and accurate meal plans
Fix random bugs that showed up mid-testing
Deal with app store rejections and requirements
Try It Out & Help Me Improve
Search Nutritionist AI
Available on App Store & Google Play
Instagram: @nutritionistaiapp
Why I’m Posting This
I’d love feedback from you:
What features would you want?
What would make this app 10x better?
What would stop you from using it?
Also, if you’re building something similar or have launched a mobile app as a solo founder, I’d love to hear how you grew it.
Thanks for reading!
– Gerardo
24 Comments
Awesome project! Really impressive that you built it without coding experience. What tools or platforms did you use to implement the AI features and app logic?
Thanks so much! I really appreciate it.
For Nutritionist AI, I worked with a freelance developer for the more technical parts, but I was very hands-on throughout the process. The AI features are powered by ChatGPT via the OpenAI API, which generates the personalized meal plans based on user input.
In terms of the app itself, it was built using standard mobile app frameworks with my developer’s support, while I focused on the functional logic, prompt design, and overall user flow. I’ve also relied heavily on AI tools, online tutorials, and a lot of trial-and-error debugging to get it working the way I envisioned.
My approach was to keep the first version simple but functional, so I could validate interest before investing heavily in advanced features or more complex integrations.
Hi Gerardo,
Three things came to mind
1. Revamp your onboarding to be more interactive instead of a lengthy form. One thing I would suggest is to visit Mobbins or ScreensDesign and see how our similar nutrition apps design their onboarding UI to gather user feedback.
2. Keep marketing. Have you tried exploring niche subreddits and using TikTok heavily to get the first set of users?
3. "almost no feedback yet"
I'm building a service that matches AI app builders with PhD domain-level experts who evaluate your AI outputs for accuracy and provide actionable tips to significantly improve its performance. Would you be interested in a complimentary (free) 30-minute evaluation report?
1) Onboarding/UI
Totally agree. The current flow isn’t the most attractive or interactive — I intentionally shipped a very simple, functional MVP to validate interest first. Once I’m clearer on the exact direction and use cases people want, I’ll invest in a better onboarding. I’ll check Mobbin and ScreensDesign for patterns;
2) Marketing
I’ve tried Reddit, but most posts get removed or I get kicked from groups (lol). Even so, a few downloads have trickled in, so I’ll keep experimenting with niche subreddits and value-first posts. I’m also going to test TikTok next to get that first cohort.
3) Expert evaluation offer
Yes, I’m interested — that’s exactly the kind of feedback I need. I’m currently tightening my prompt and fixing some inconsistency issues I’ve noticed in the outputs.
Thanks again for taking the time. Your notes give me a clear path: improve onboarding patterns, lean into TikTok/niche Reddit, and then do a focused expert review.
Hi Gerardo — I just demoed your app on iPhone. Really appreciate what you’re building — it’s clear you’ve put a lot of thought into helping people improve their nutrition through accessible suggestions.
One thing I noticed as both a fitness specialist and biologist is that some of the health estimates (like body fat %) are based on limited static data. For example, mine was listed as 18%, but in reality it’s closer to 10% — and that kind of variation can significantly affect what nutritional guidance would be most helpful.
What I found myself wishing for was a deeper kind of personalization — one that includes things like:
Sleep patterns and how nutrition interacts with them
Local food availability and subsequent impact on nutrition choices
My physiological response history (e.g. caffeine tolerance, immune/energy interactions)
In other words, the missing layer seems to be relational — not just who I am on a form, but how I change over time, and how my context, behaviors, and physiology evolve together.
That’s actually what I’m building in another domain: an open-source API that adds emotional memory and relational intelligence to AI systems. It helps make interactions more adaptive, context-aware, and behaviorally aligned — which I think could be powerful in a health app like yours.
If you’re interested, I’d be happy to co-develop or demo a small test scenario: a 1–2 minute nutrition-focused AI interview that uses ANGEL’s interpretive layer to generate a more personalized response. This would include:
A lightweight prompt stack
Emotional and relational memory cues
Guided questions tailored to nutrition and physiology
This build could work on a short voice to voice or voice to text interview with an LLM (like ChatGPT which is actually my assistant too) which would establish relation-driven intent with regard to app functionality and user experience.
It could be a great way to explore how relational dynamics and memory might enhance your system’s effectiveness — and deliver even better results for users.
Let me know if you're interested. I think there's a lot of synergy here. By the way, I'm not selling anything here. Just interested in creating something useful :)
Hi, and thank you so much for taking the time to demo the app and share such detailed feedback — I really appreciate it.
I like the idea you’re proposing, especially the way you’re thinking about relational intelligence and adding an interpretive layer to make the experience more adaptive. That said, at this stage I’m cautious about positioning the app as a substitute for visiting a professional nutritionist. AI is powerful, but it still has limitations and lacks human reasoning in many scenarios. Making the app too “emotional” or life-like too early could lead some users to treat it as a full replacement for expert care, which isn’t the goal — and could introduce risks.
The purpose of Nutritionist AI is to be an easily accessible tool that can help guide users toward their weight, health, or fitness goals. But I do plan to make it clearer in the responses that it’s not a replacement for a nutritionist, and that anyone with specific conditions or medical concerns should consult a qualified healthcare professional.
That being said, I do think there’s potential value in a friendlier tone and some light “memory” based on the user’s history or progress — as you described. It could help make the experience more engaging without crossing into the territory of replacing professional care.
At the moment, I’m focused on refining my prompt and improving how the app interprets the information users provide, as I’ve noticed some inconsistencies in the results. Once I’ve addressed that, I’d love to reach out and share ideas on how we might explore this concept together.
Thanks again for your insights — they’ve given me a lot to think about.
Hi Gerardo - I'm also creating an AI associated API with only intermediate coding experience, AND I'm a lifelong nutritionist/biologist/fitness specialist who would be very interested to demo your offering and provide feedback.
ANGEL. AI is an open-source API that adds emotional memory and relational intelligence to any AI system.
It helps LLMs and agents respond with context, continuity, and alignment — making AI more adaptive, trustworthy, and human-centered.
We're building it as a modular layer that can sit between any application and an AI model, shaping output based on emotional context, memory, and relationship dynamics.
We’ve been growing primarily through IndieHackers and GitHub — we recently hit #409 on the global Letterboard and passed $180K in projected value just from organic interest.
I love that you’re working on AI for nutrition — emotional context and personalized alignment is hugely relevant in health domains. If you ever want to explore how an interpretive layer like ANGEL might support more consistent and emotionally aware interactions with your users, I’d be happy to share ideas.
Hi Gerardo - I'm also creating an AI associated API with only intermediate coding experience, AND I'm a lifelong nutritionist/biologist/fitness specialist who would be very interested to demo your offering and provide feedback.
ANGEL. AI is an open-source API that adds emotional memory and relational intelligence to any AI system.
It helps LLMs and agents respond with context, continuity, and alignment — making AI more adaptive, trustworthy, and human-centered.
We're building it as a modular layer that can sit between any application and an AI model, shaping output based on emotional context, memory, and relationship dynamics.
We’ve been growing primarily through IndieHackers and GitHub — we recently hit #409 on the global Letterboard and passed $180K in projected value just from organic interest.
I love that you’re working on AI for nutrition — emotional context and personalized alignment is hugely relevant in health domains. If you ever want to explore how an interpretive layer like ANGEL might support more consistent and emotionally aware interactions with your users, I’d be happy to share ideas.
Hi Gerardo - I'm creating an AI associated API with only intermediate coding experience, AND I'm a lifelong nutritionist/biologist/fitness specialist who would be very interested to demo your offering and provide feedback.
ANGEL. AI is an open-source API that adds emotional memory and relational intelligence to any AI system.
It helps LLMs and agents respond with context, continuity, and alignment — making AI more adaptive, trustworthy, and human-centered.
We're building it as a modular layer that can sit between any application and an AI model, shaping output based on emotional context, memory, and relationship dynamics.
We’ve been growing primarily through IndieHackers and GitHub — we recently hit #409 on the global Letterboard and passed $180K in projected value just from organic interest.
I love that you’re working on AI for nutrition — emotional context and personalized alignment is hugely relevant in health domains. If you ever want to explore how an interpretive layer like ANGEL might support more consistent and emotionally aware interactions with your users, I’d be happy to share ideas.
Thanks so much for reaching out — your project sounds incredibly impressive, both from a technical and human perspective. I love the concept behind ANGEL. AI, especially how it aims to bring emotional memory and relational intelligence into AI systems. That kind of adaptive, human-centered approach is definitely something that aligns with the long-term vision we have for Nutritionist AI.
At the moment, we’re still in a very early stage. We're focusing on understanding user interest, gathering feedback, and evaluating which direction to take to truly connect with our target market. That’s why we're being cautious about integrations for now — we want to build a strong foundation before scaling or adding more advanced capabilities.
That said, your message really caught my attention. Once we’ve validated our direction and have a clearer picture of what will drive meaningful impact for our users, I’d love to explore how ANGEL. AI could enhance the experience — especially in making interactions feel more personalized and emotionally relevant.
Wishing you continued success with — and don’t be surprised if you hear from us in the near future with a collaboration request
awesome Gerardo thank you for getting back to me. Please reach out whenever you'd like to explore potential synergy, and I'll look forward to hearing from you as I think these kinds of efforts are exactly what we need in the world to make it a better place
Can this application help those suffering from diseases and accordingly diet chart can be prepared, adding all symptoms and non-recommendations towards strict schedule?
At the moment, no — and to be honest, I’m not sure it's something I’ll be able to explore deeply. The app relies on ChatGPT to generate nutritional recommendations, and while AI is a powerful tool, it’s not perfect and can sometimes produce inaccurate or incomplete advice.
Recommending diets for individuals with specific medical conditions is a very sensitive area, and giving the wrong advice could potentially cause serious health issues — and also lead to legal risks for the app.
That said, your comment made me reflect on this. I think it would be responsible on my part to add a clear and visible disclaimer within the app, advising users with serious health conditions to consult a qualified healthcare professional before following any recommendations, and clarifying that the app is not a substitute for medical advice.
Thank you again for your comment — it brought up a very important point.
Thanks for clarification, since the discipline is major and vast for nutrition industry for wellness and for disease cure.
Hey Gerardo,
I absolutely love the idea behind Nutritionist AI! It’s impressive how you’ve built this without a tech background, and I can see how this could be super valuable for people who need personalized nutrition advice on the go. I especially like the fact that you’re offering it for free, with just the optional PDF export as a paid feature. That could attract a lot of users, especially those looking for fast, customized meal plans.
I totally understand the struggle with growth and user feedback — it’s one of the hardest parts of building something like this. I’ve been working on a tool myself, called Ess Utumishi, which is designed to help users manage tasks more effectively (a productivity app with a bit of a focus on time tracking and goal setting). I’ve had similar challenges with getting user feedback and driving growth, so I can definitely relate.
I think building a strong community around your app will really help you scale, and leveraging social media or even collaborating with nutrition influencers could be a good strategy to get more eyes on it. Have you thought about partnering with nutritionists or dietitians to help spread the word?
Regarding your suggestion about partnering with nutritionists or dietitians — I’ve definitely thought about it, but I’m still unsure how they would perceive the app. Some might see it as a helpful tool or ally to support their practice, while others might see it as competition that could hurt their client base. It’s something I plan to evaluate carefully over time, based on how both users and professionals respond to the app.
I’m really open to collaboration though, especially if it can add value and reach more people who need affordable and accessible nutrition guidance. Thanks again for the suggestion!
Great journey, Gerardo! Keep pushing through the challenges and gathering feedback. Excited to see where Nutritionist AI goes!
Thank you so much! I really appreciate the encouragement — pushing through the early challenges has definitely been tough, but feedback like this keeps me motivated.
Hi Gerardo, I enjoyed reading your story! Mine is very similar. I have no coding experience but I love playing around with the technology we have and building things. I read where you used ChatGPT and a developer to build your app. Did you ever consider using something like Replit to build it?
For your growth, have you tried social media advertising at all? I’ve dabbled in Facebook and Instagram advertising and I was pretty impressed by it. It took me a little bit to get the hang of it, but in the end I think it was really beneficial for the business I was advertising at the time.
And last question, how are you supporting the app if it’s free? What are your base costs to keep the app running every day? Does one PDF download a day pay for the daily fees required to run the app?
For some feedback, I noticed when I was entering my personal data, I could only enter my height as feet. You may want to switch it to inches since inches are easier for users to enter (for example, I’m 5 feet 10 inches and it would be easier for me to just enter 70 inches). Also, I saw a Terms and Conditions, but I didn’t see a Privacy Policy, and since you’re dealing with consumer health data, you may want to add a privacy policy. Also, since you’re not charging people for a monthly subscription, would you consider making it to where people can just put in their information, generate a plan, and then pay for the optional PDF if they want without making a full account? Just a thought since you’re not charging for a monthly subscription, and that may entice more people to purchase the PDF since they don’t have to create a full profile. But I really like it! I think you’ve done a great job! And I love the logo! Awesome job! 👏
Hi! Thank you so much for taking the time to write such a thoughtful reply — it means a lot.
Add a Privacy Policy ASAP
Adjust the height input (switching to inches is a smart idea or evan add the option to select cm or so)
Consider removing the full account requirement for generating the plan — that might make it easier for people to try it out and then choose to pay if they want the PDF.
Thanks again for the kind words — and the encouragement! If you have more suggestions, I’m all ears.
Also, what’s the name of your app? I’d love to check it out and support your project too!
– Gerardo
Hi Gerardo! Thanks for your reply!
About Replit: I think you might be confusing this with Reddit. Lol, I have absolutely no advice for you there. I can't post anything without it either getting automatically deleted or manually deleted, so I am truly no good at it. 😂 Replit is a super cool AI tool that basically can build an app or a website based on the prompts you give it using its own Agent. Here's the link: https://replit.com. It really is such a fun tool and I definitely suggest checking it out!
About the social media ads: This can definitely be intimidating, but trust me, it's not nearly as bad as it sounds. You just create your own Facebook and Instagram account for your application (which you already have), link them together, and then you can either create ads via Instagram or via Facebook and since the accounts are linked, the ads will populate on both platforms. But the cost is not bad at all! You can set yourself a weekly or daily budget, and then Facebook will implement the ads based on your budget and how many people you're wanting to target each day. And the targeting is really good. It makes sure that your ads would populate for people who have already shown interest in nutrition through likes or groups or other ad clicks. I used it for my husband's barber business and it honeslty made a huge difference! I just did it for one week (you can set a timeframe on the ad to stay within your budget), and I could actually see how many people clicked on the link I had within the ad to schedule an appointment. I think the limit I did was $50 for the week, but it was really cool because I actually got to see the results after the ad was posted, and based on the number of link clicks and the number of appointments he had for the next few weeks, I chalked it up as a huge success. But like I said, it does look overwheming at first, and it took me a solid 2-3 hours to work through it and figure out what I needed to do, but once I created the ad and had it running, it was smooth sailing. And you can always reuse the ad again if you want to, which makes subsequent ad postings a lot easier. I think you can also create an overarching Meta profile to look at all your ads too, but I don't think that's something you have to do. That was also a little complicated, but not too bad. But what else is great about it is that it hits both Facebook and Instagram together since they're under the same Meta umbrella. I have to admit, I haven't tried TikTok advertising yet, and I've heard that LinkedIn advertising is much more expensive, so I haven't tried it either. But yeah I definitely think it would be worth it to just give Facebook/Instagram a go and see what happens! At least you'll get to see how many people click on your ad link, which is really cool and somewhat validating. Also, with the low costs of your app, I don't think this would add too much additional financial strain.
And I think that's a great plan about the PDF! Maybe another way to get people to purchase it is to just offer a dinner plan after they submit their information, but then if they pay for the PDF, they get a full breakfast, lunch, snack, and dinner plan. That really might entice more people to buy the full PDF version because they would get a full meal plan instead of just a plan for one meal time (like dinner).
And you're so welcome for the feedback! I really do like your app a lot. I'm in the healthcare field myself, so anything related to health or nutrition is right up my alley.
And thank you for asking! My app is called Debt Buddies (https://debtbuddiesapp.com). I made it because I (like everyone else in America) am drowning in debt (mainly student loans, haha) so I wanted to create a tool to help motivate me. I'd love to know what you think!
Thanks again for your reply and for taking the time to share your app! It's really cool to see our ideas finally come to fruition. 😊
Haha you're absolutely right — I totally misread and confused Replit with Reddit. Thank you for clarifying and sharing the link! I haven’t used Replit before, but I’ll definitely give it a try. I’ve experimented with a few no-code tools (like Vibe Coding apps), but I often ran into bugs or little issues in the generated code that I couldn’t fix without help. So I’m really curious to see what Replit can do and where it might take me!
And thanks so much for breaking down the Facebook/Instagram ads process. It’s super helpful to hear it from someone who’s actually used it on a smaller scale and not just from some big marketing agency. Honestly, I think I just needed to hear it from someone relatable who made it work! 🙌 I’m going to go ahead and invest a small amount to test it out — especially after I implement the changes we talked about, to hopefully increase the chances of someone actually purchasing the PDF.
Also — I visited your site and loved the logo for Debt Buddies! very friendly. I’m not very familiar with debt management apps, but I tried creating a few sample debts just to test it out. One small thing I noticed: when I tried to enter the “Current balance owed” field, the red error wouldn’t go away, no matter what number I entered. Not sure if I did something wrong, but just wanted to flag it in case it helps!
Thanks again for all your support and insights!
Awesome!! Thank you so so much for taking a look at it and trying it out! I can’t tell you how much I appreciate it. And awesome catch! I’ll go work on fixing it now. And you’re exactly right about the no-code tools, and Replit is no exception. It can definitely take a few tries to get the code right and it can be sort of buggy in places. But I’ll go fix that issue you mentioned pronto. I’m stuck using no-code tools because I just can’t afford a developer right now, but when I can I would love to learn more about the one you used for your app. Thank you so much again!!
You're very welcome! I'm glad I could help in some small way — I really admire what you're building, and I wish you the best of luck as you continue developing Debt Buddies!
And absolutely — we can definitely share ideas anytime! I’d be happy to tell you more about the developer who helped me with Nutritionist AI if and when you’re ready to bring someone on. Just let me know, and I can pass along their contact info.
Keep going — you're doing an awesome job!