The Food Plan I Needed But Couldn’t Find
0x41434f
I’ve been thinking about this idea for a long time. Not because I wanted to build an app, but because I hit a moment in my life where I realized I needed something that didn’t exist. I’m a cyclist, swimmer, and walker. I take it seriously. I train hard and track everything. I log my rides with Strava and monitor my sleep, weight, and recovery with Fitbit. I like understanding how my body responds to effort, stress, and rest. I’ve been doing this for a while because it helps me understand myself. It gives me something to compare against when I’m out on a ride, in the pool, or even on a long walk.
Even with all that data, there was one part of my routine that never felt solid. Food. I could tell when I ate wrong. I could feel it on the bike or the next morning. Some days I felt strong and light. Others I felt slow or under-fueled. But there was no system helping me make food decisions based on how my body was actually doing. Everything felt disconnected. I was guessing. I had apps that tracked calories or logged meals, but none of them cared about what I had done yesterday or how I slept last night.
Then I went in for routine bloodwork and found out my LDL cholesterol was high. That completely shifted the way I thought about food and performance. I hadn’t been testing for low-density lipoprotein cholesterol. I was there for something else. But the result caught me off guard. I’m lean. I train regularly. I thought I was doing everything right. But I had been eating a lot of chin-chin and plantain chips. Snacks I grew up with and still enjoy. I just didn’t think they were doing that much harm.
That was the moment I knew something was missing. I had the discipline to train. I had the wearables. I had the data from my lab results. But I didn’t have a system that looked at all of that information and helped me decide what to eat. Not just in general, but today. Based on how I slept, how I trained, my weight trend, and my health concerns and turn all of it into something clear and actionable. A daily plan that told me what to eat, what to avoid, and what to buy.
This idea isn’t built yet. What I have is a map. I’ve written the flows, created the mockups, explored the Fitbit and Strava connections, and outlined how the system should behave. But I haven’t written full code or shipped a real app. I’ve only started shaping the version of this system that I wish had existed when I saw those lab results.
The way I see it, here’s how it would work. You connect your Fitbit. You connect your Strava. The system checks your sleep, your heart rate, your weight trend, and your workout history. It notices whether you’ve been pushing hard, recovering slowly, or trending in the wrong direction. Based on that, it gives you a daily or weekly meal plan tailored to where you are right now.
If you’ve been sleeping poorly, the meals shift to something lighter and easier on digestion. If your training has been intense, it increases your calories and nutrients to support recovery. If you’re showing signs of overtraining, it cools things down. If your lab results show high cholesterol, it removes saturated fats and adds more fiber. If your weight is dropping too fast, it would suggest higher-calorie meals with nutrient-dense ingredients. If it’s climbing and you’re trying to lean out, it guides you. The system doesn’t just count calories. It uses signals from your body to suggest meals that match your condition, activity level, and goals.
Then it builds a grocery list. Clean, categorized, and easy to follow. You can check off items, rearrange them, and export them as a PDF or CSV file. You can even select your store, like Walmart or Sam’s Club, and have the export reflect that. The system removes friction at every step. It helps you go from plan to plate with as little thinking as possible.
But it doesn’t stop there. The system would also allow you to select dietary goals and restrictions. You could choose low cholesterol, diabetic-friendly, low sodium, high protein, keto, Whole30, paleo, gluten free, or plant-based. It would respect those settings and remove ingredients that don’t align. These aren’t just toggles. They shape the actual structure of the plan. The system could even support clinical considerations, like meals for people on GLP-1 medication, kidney-friendly plans by stage, heart health diets, or plans designed for Crohn’s or GERD.
The reason this matters is because most people don’t realize when food starts working against them. I didn’t. Until that lab result. It’s easy to assume that if you’re training and tracking, you’re on top of your health. But food is where things slip. And the systems we use today don’t bring all the right data together. They track, but they don’t guide. They report, but they don’t help.
That’s where this system fits. I’ve thought about how to layer in expert oversight. Chefs could review the AI-generated plans and adjust them. Fix weird combinations, add cooking tips, and improve the meals without changing the logic. Plans could be labeled “Chef Reviewed” for quality. The same goes for licensed dietitians. They could review a user’s plan and data, then make slight edits based on lab trends or user preferences.
Later on, this could also integrate with commercial kitchens. A user could export their grocery list or send it to a prep partner to get real food shipped to their door. This makes the system useful for both home cooks and people who just want the meals handled for them.
But even without kitchens or chefs, the core value stays the same. A system that reads how you slept, how you trained, what your goals are, what your labs say, and gives you a clear, simple food plan. A plan that you don’t have to think twice about. One that supports your performance, your recovery, or your health, depending on what matters most to you.
I know pro cyclists already have a version of this. They don’t guess what to eat. They have nutritionists who work with coaches to build exact food plans based on training blocks, race calendars, and recovery. But even then, I imagine a lot of that planning is static. Manually adjusted. Not responsive to daily health data. Not something that adapts on the fly based on wearable signals or bloodwork. And definitely not something most people have access to.
What I want to build is a version of that system that anyone can use. A system that starts with your data and your body. A plan that reflects how you live, how you move, and what your health is trying to tell you. A daily rhythm that connects your training with your food, without needing a full team behind you.
This is not a polished product. It’s not a polished pitch. It’s the system I’ve been building in pieces. In mockups. In notes. In the way I already track my own life and try to fix the gaps I feel.
This isn’t just another app. It’s the thing I needed when I got that lab result. When I was trying to make better choices but didn’t know where to start. When I was trying to turn data into direction. This is what I wish had existed for me. And if no one else is building it, then I will.
Development & Design Note #1 (April 11, 2025):
After writing about this idea, I have started building a very early version of it using AI coding tools (like the "vibe coding" I mentioned in another post). The demo screens included here show this first step.
