Aisle1

Inspiration

The spark for Aisle1 came from a personal frustration with existing health apps like Yuka. While Yuka is great for scanning physical items in a store, it’s incredibly clunky for online grocery shopping. I found myself sitting at my laptop, manually typing product names into my phone just to find a barcode to scan. It was a slow, multi-step process for every single item in my cart. I realized that if I wanted to eat healthier without spending three hours on a 10-minute grocery run, I needed a solution that lived where I shopped: the browser. I wanted a "one-click" experience that leveraged AI to do the heavy lifting for me.

What it does

Aisle1 is a smart grocery co-pilot that lives in your browser. As you browse sites like Amazon, Walmart, or Kroger, the extension automatically detects the product on your screen, pulls its real-time nutritional data from the USDA database, and calculates an unbiased health score. Instead of just telling you a product is "bad," Aisle1 acts as a helpful guide—suggesting three data-verified, healthier alternatives directly in a sleek sidebar. We’re moving the nutrition label from the back of the box to the front of your digital decision-making.

How we built it

We built Aisle1 using a modular Manifest v3 architecture. The frontend is a lightweight Chrome extension that handles product detection through DOM scraping and metadata extraction. We offloaded the heavy lifting to a custom Node.js API backend. The extension sends the product metadata to our API, which serves as the brain of the project—querying the USDA FoodData Central and SerpAPI for alternatives. By centralizing the logic in an API rather than the extension itself, we ensured the tool remains fast, responsive, and easy to update without forcing users to reinstall the extension every time we tweak the scoring algorithm.

Challenges we ran into

Our biggest hurdle was the Nutrition Scoring Logic. We quickly learned that "healthy" isn't a black-and-white metric. Our early versions were either too harsh—flagging almost everything as unhealthy—or too soft, failing to catch hidden additives. Achieving consistency across different food categories (like comparing a protein bar to a bag of spinach) required us to rewrite the scoring algorithm multiple times. We had to balance USDA raw data with processing-level classifications to ensure that our "Global Health Score" was both accurate and fair.

Accomplishments that we're proud of

We are incredibly proud of our Product Detection Engine. Building a script that can accurately identify a specific grocery item across vastly different website layouts from the cluttered pages of Amazon to the minimalist UI of Whole Foods was a major win. Additionally, successfully bridging the gap between raw USDA data and a user-friendly "NutriScore" interface was a huge milestone for us. We turned complex biochemical data into a simple, actionable insight that anyone can understand in seconds.

What we learned

This project taught us that UX is just as important as Data. You can have the best health data in the world, but if it takes more than two seconds for a user to see it, they won't use it. We also gained deep experience in API design and data normalization, learning how to take "messy" web data and turn it into a structured format that a machine can score reliably. Above all, we learned that building for "transparency" requires a high level of technical ethics—ensuring our scores remained unbiased was our top priority.

What's next for Aisle1

We’re just getting started. Our roadmap includes:

Platform Expansion: Extending support to every major regional grocery chain globally.

The "Universal Link" App: Developing a mobile web app where users can simply paste a product URL to get a score, removing the "barcode paywall" seen in competing apps.

Transparent Monetization: We want to keep Aisle1 ad-free. Our goal is to implement a "Honey-style" affiliate model where we suggest healthier swaps. If a user chooses a better alternative through our link, we earn a small commission—aligning our success with the user’s health.

Built With

• fastify
• javascript
• manifest
• nutri-score
• serperapi
• tailwind
• usda-nutrition-data