1) Former Chief Data Engineer of Paypal (also was at Yahoo and mentor at Apple/Google and startups): Did a voice enabled app startup. Generally doing Data Science and Engineering including Natural Language Processing and Chatbots. Participant and frequent winner of many hands on Hackathons. Full Stack Software Developer. See http://linkedin.com/in/youssefi
2) Former Head of Analytics of Instagram and PM for Facebook Chat (also PM at Uber, Google, Yahoo and a startup): Product Design, patient/subject matter expert and Mobile Developer. See https://www.linkedin.com/in/anvari/
(we have been friends, classmates and co-worker for 20+ years)
3) Advisor: Head of Google Personal Assistant
As a person dealing with diabetes for years, I find it challenging to monitor my food intake and predict what and how much I can eat or drink next instantly. This, turns out to be a common problem for the diabetes community.
How Zee App is built and operated
Makes personalized recommendation on what to eat/drink next and how much based on Machine Learning algorithms and crowd-sourced data. Makes it easy for me to record a diary of food/pill intake via short voice recordings on my phone which is basis of personal recommendations tailored to my body and metabolism. It also does helpful services such as ~2 hour reminder after each meal (when glucose is highest) for the patient to consider measuring the blood glucose, calorie counting and portion deduction from food photos using Deep Learning (CNN) for Image Recognition and saving photos in dieries. It can potentially show portions and calories in AR (augmented reality) as you point the phone camera at food.
How it is built
Zee is a cloud service connected to mobile app and generic Blood Glucose Monitoring hardware device. Main innovations are algorithms, data and integrations.
Income is from subscription for value-added services. Cost of building the cloud and mobile app are moderate and sustainable.
Current state of the art NLP, Machine Learning and Recommender Systems are quite capable of addressing the basic needs. Over time we gather a lot of crowd sources data that enables Deep Learning algorithms which need a lot of data.
Any company interested in the space such as Google which builds Glucose Monitoring eye lenses would be interested in acquiring the company for the system, team and data.
Team is of exceptional relevance and strength.
To our knowledge there is no diabetes focused personalized, mobile, voice enabled device with this level of data and algorithmic complexity and ease of use. With ease of access to smart phones and mobile internet this is accessible big parts of diabetes community can benefit from Zee's personalized recommendations. Data gathered from many people can be used anonymously to improve outcome.