This post is my submission for DEV Education Track: Build Apps with Google AI Studio.
What I Built
I built Healthy Meal Visualizer & Nutrition Tracker. The app lets you input a set of food ingredients (or select from suggestions), then Gemini creates a balanced recipe + full nutritional breakdown (calories, macros, etc.), and Imagen generates a photorealistic image of how the meal might look when plated.
Key prompts:
- Using [ingredients list], create a healthy recipe under [target calories], with macros and portion sizes.
- Generate a fresh, appetizing food photo of the dish using those ingredients, in natural lighting.
Demo
My Experience
- Learned how to adjust prompts for Imagen to get realistic food plating & lighting, and avoid weird artifacts.
- Found that nutrition data alignment (ingredient names, portion sizes) needed careful handling to avoid wrong macro calculations.
- Surprised by how well the image looked once I tuned the prompt. But also saw that sometimes image generation makes the meal look more perfect than likely in reality.
- If I did this again, I’d try integrating image recognition for user-uploaded photos, and personalize suggestions for dietary restrictions.
Top comments (1)
This is actually a cool idea. Seeing meals laid out visually makes healthy eating feel less abstract and more doable, especially for people who get stuck on what to cook. I’ve found that tools like this help with portion ideas and variety, even if you don’t follow recipes exactly.