Tales From Startup Alley: Klappo and Big Food Data

My Startup Alley blogging is a little delayed this year. My apologies to those who are still interested in the TechCrunch Disrupt event from two whole weeks ago. So London-based Klappo was one of the startups I regret not seeing in person. They describe themselves as “connecting people with ingredients”, and at TC Disrupt they were serving up new APIs to allow for more creative foodware.

The Klappo team has an enormous database of food knowledge: over 2 billion data points involving recipes, ingredients, and product bar codes, according to their website copy. At TC Disrupt they introduced their Sensum.io APIs for allowing developers to tap into this data. Klappo provides more than just a straight lookup ’cause they’re boasting semantic knowledge of food.

Once upon a time, I got a taste of semantic programming playing around with Freebase‘s fabulous APIs. By the way, they’re now part of Google, helping to make their search results far smarter.

So what is Klappo about? Maybe their Venn diagram is the intersection of Big Data, Machine Learning, and Mario Batali. I took a peek at their APIs documentation, which are a techno-foodie’s dream come true: you can parse natural language recipes and pull out the fundamental food objects as a JSON object. There’s another API that looks at the parsed ingredients to ensure they fit into a dietary regime–Paleo, gluten-free, etc.

I’m thinking out loud how this would greatly improve some of my go-to recipe sites–Food52, Smitten Kitchen, and Epicurious.

Their goal, based on a brief email chat with them, is to focus on health use cases–hospitals, medical devices and apps–but the options are pretty wide open. Besides making web-based recipe sites a lot more powerful, there’s an idea in their marketing collateral for supermarket chains to provide really detailed information on their products. I’m picturing a food info kiosk powered by Klappo in, say, Whole Foods.

Yes, I signed up for a beta version of their APIs. Let you know what I find out.