In a world where food labels hide more than they reveal, consumers are left guessing about what they’re really eating. Over a 13-week capstone project at Langara College, I joined a multidisciplinary team of designers and developers to create a game-changing solution.

PeekaBite is an AI-powered mobile app that scans packaged foods, decodes nutritional facts, flags hidden health risks, and recommends healthier alternatives. Our goal was to transform food label confusion into actionable clarity, empowering users to take control of their health one scan at a time.

I check labels all the time, mostly for sugar and sodium content, but it’s hard to figure out what’s really healthy.

- Participant 3

Despite good intentions, users often struggle to make healthy decisions due to:

• Overwhelming labels: Technical terms and serving sizes make food labels hard to decode.
• Time-consuming comparisons: Grocery shopping is fast-paced there’s no time to evaluate multiple products.
• Hidden health risks: Marketing often obscures critical details like harmful additives, hygiene standards, or production processes.

We began by conducting user interviews and reviewing health studies. We also explored how cosmetics apps highlight harmful ingredients and realized the food industry lacked such transparency.

• Key Research Themes Health Scoring: Users want an easy “yes or no” for food choices.
• Additive Warnings: They care about what's not shown on the label.
• Deeper Transparency: From packaging to brand ethics users want to know more.

‍

To better understand our audience, we created a persona that captures their goals, frustrations, and habits.

‍

‍

‍

‍

The “Oatly” Revelation

While researching, we discovered Oatly’s controversies regarding additives and misleading branding. This real-world example proved our users' concerns were valid and inspired our most powerful feature: the AI Report, a comprehensive breakdown of hidden product details.

Competitor analysis

Based on our findings, we pinpointed the unique features that distinguish PeekaBite from other similar products.

‍

‍

‍

PeekaBite is built around three key features:

• Product scanning and Label interpretation: Barcode or label scan gives immediate feedback.
• Personalized Health Scores & Alternatives: AI customizes health scores based on user goals.
• Peeka AI Report: A breakdown of sanitary risks, banned ingredients, packaging concerns, and brand history.

Usability Testing Outcomes Health Scores: Well received, but needed clearer color coding and labels. Additive Alerts: Users asked for visual indicators (e.g., red/yellow/green icons). AI Report: Users appreciated depth but preferred shorter summaries with toggle options. Bonus Feature: Chat With Peeka Our dev team implemented an interactive mascot-powered chatroom. We refined this with default question prompts for efficiency and reduced effort.

‍

Challenge: Incomplete Product Data

When a scanned product wasn’t in our database (sourced via OpenFoodFacts), users had to manually input product info leading to quality control issues.

Our Fix: We added a Report Feature that let users flag incorrect data. This empowered the community while safeguarding our database accuracy.

To reduce feature creep and simplify development, we focused on: 3 Core Features: Health score, alternative suggestions, AI Report. Optimized Flow: Scanning → Score → Deeper insights or swap. Scalable Design: Built a foundation for future feature integrations.