Case Studies
“Tribe has been a great partner and their model works exceptionally well. I’ve loved the flexibility and having a demo every week that we can share with the rest of the company.”
Dustin Manderfeld, VP of Data Science & Engineering
Opportunity
MyFitnessPal helps more than 280 million people track their nutrition against a database of more than 20 million foods.
That scale makes a simple product idea technically difficult. MyFitnessPal wanted members to describe a meal—or an entire day of eating—in everyday language, then turn that description into accurate, structured food entries.
The challenge wasn’t just capturing speech. It was connecting unstructured descriptions, quantities, brands, and serving sizes to the right records in MyFitnessPal’s food database.
MyFitnessPal brought in Tribe to work alongside its product organization. A team of AI engineers and product managers scoped the experience, built a working prototype, and iterated through weekly demos with stakeholders.
"I want to be able to tap and speak all of the ingredients or food items I ate for a given meal so I don’t have to search and type each one individually. Or multiple meals—if I can say everything I’ve eaten for the entire day, why can’t I log all my food in one step?" — Dustin Manderfeld, VP of Data Science & Engineering, MyFitnessPal
What we built
By the end of the sprint, the team had built a functional Voice Log experience capable of capturing multiple food items in a single interaction.
Members describe what they ate in everyday language and receive suggested matches from MyFitnessPal’s food database. They can review the results, adjust serving sizes, quantities, or meal details, and add the complete entry to their diary.
Powered by Claude 3.5 Sonnet on Amazon Bedrock, Voice Log turns conversational input into structured food data, including items, quantities, serving units, brands, and meal context. MyFitnessPal then matches that data against its food database.
Voice Log is now in production for MyFitnessPal Premium users on iOS and Android.
The feature turns what previously required multiple searches and individual entries into one conversational flow. Members can name several foods in a single phrase, review the suggested matches, make adjustments, and add the results to their diary.
The work also gave MyFitnessPal a repeatable way to prototype AI product ideas. Leadership established a dedicated prototyping environment and expanded the company’s culture of experimentation—creating a faster path from idea to working software.
Powered by Claude, Voice Log combines Anthropic’s model intelligence with MyFitnessPal’s product knowledge and food data. Tribe’s team carried that combination from sprint to production.
