Eat this, exercise now; new personalized software predicts and helps prevent blood sugar spikes

Not all have type 2 diabetes, which induces consistently elevated blood sugar levels, but many do.

Around 9% of Americans suffer and another 30% are at risk of contracting it.

Enter a four-year-old, subscription startup, January AI, app that started providing its customers with customized nutritional and activity-related recommendations in November, based on the combination of food-related data which the company collected secretly over three years and on the unique profile of each user over the first four days of the software.

Why the need to customize?

Since people believe or not, any food, from rice to salad dressing, can be reacted very differently.

Tech can sound worldly, but it's eye opener, potentially live-saving, promises co-founder and CEO Noosheen Hashemi and her co-founder, Michael Snyder, a Stanford genetics professor who has been dealing with diabetes and pre-diabetes for years.

Investors really like the proposal.

Felicis Ventures has contributed to a seed funding of 8,8 million dollars, with HAND Capital and Salesforce founder Marc Benioff joining the firm.

"While other companies have made headway in understanding biometric sensor data—from heart rate and glucose monitors, for example—January AI has made progress in analyzing and predicting the effects of food consumption itself [which is] key to addressing chronic disease."

To hear more, we spoke to Hashemi and Snyder this afternoon who raised a total of 21 million dollars.

Underneath is part of our conversation, condensed for clarification and length.

TC: What have you constructed?

NH: We have developed a multi-purpose framework where we gather data from multiple sources and forecast people's glycemic reaction to their choices.

We take evidence from cardiac rates tracking and continuous glucose monitoring as well as a 1,000-person clinical trial and a 16 million food atlas, for which we have extracted nutritional values and created nutritional markings through machine learning.

[The idea is] to determine what your glycemic reaction would be to any food in our database after a four-day preparation.

They don't have to eat the food to know whether they should eat it or not; our product gives them their choice.

TC: There has also been glucose testing before, although this is predictive.

Why does this matter?

NH: We want to consume the love and erase the shame.

For eg, we can determine how long after you consume some food in our database to keep your blood sugar at the right amount.

We don't want to tell you what to do about it; we want to tell you what to do about it.

We will tell you if you are dreaming of fried chicken and a shake: you will have to exercise 46 minutes later in order to retain a safe range of [blood sugar].

Do you want to do the uptime?

No? No?

So maybe on a Saturday [eat the chicken and shake].

TC: This is subscription app which works with other wearables for a duration of three months, costing $488.

NH: That's retail pricing, so we've got an initial $288 offer.

TC: Do you think about customers buying this product, having a sense of what they should do better, and then terminating their subscription?

NH: No. NH: No.

Pregnancy changes, age changes, [one's profile].

People fly and don't all eat the same things.

MS: I have been wearing wearables for seven years [continuous glucose surveillance], and I still learn things.

You immediately remember that you spike through the roof, for example, every time you eat white rice.

For certain people, that's real.

But we will also be selling a year-long subscription soon, as we know that occasionally people slip [just to remember] that these boosters are really useful.

TC: How technically does it work?

Tell me, I'm in a restaurant and I'm in a pizza mood, but I don't know who to order.

NH: The curve over curve can be compared to see what is healthier.

You will see how far [depending on the surroundings] you have to walk.

TC: Do I have to chat to my mobile all of these toppings?

NH: January is checking barcodes, recognizing images as well.

It also has a manual entry and [commands] are required.

TC: Are you doing something else that you have aggregated and that you enrich with your own data with this big food database?

NH: We certainly won't sell personal details.

TC: Results not yet aggregated?

It sounds like a helpful directory.

MS: We're not 23andMe; that's not the target, actually.

TC: You said that somebody's blood sugar would increase, which is incredible.

What are some of the stuff about what your software can reveal to people?

NH: The way people react to glycemia is so distinctive, not just from Connie and Mike, but also from Connie and Connie.

If you eat nine days a row, the glycemic reaction could vary every nine days regardless of how much you slept or felt the day before and how much fiber was in your body and what you were eating before you slept.

Action is necessary before eating and after eating.

Fiber is important. Fiber is important.

It is the most ignored American diet interference.

Our ancestral diets produce 150 grams of fiber a day; today's typical American diet is 15 grams of fiber.

Many health conditions may be caused by a shortage of fiber.

TC: It seems like coaching with your app will be useful.

Is a coaching part available?

NH: Today we don't give a coaching aspect but we're in contact with many coaching solutions to be their AI partner.

TC: Who else are you with?

Companies of healthcare?

Employers who will benefit from this?

NH: We market directly to customers, but for two years we still have a pharmaceutical client.

Pharmaceutical firms are keen to partner with us as we will use lifestyle as a biomarker.

We give you [anonymized] visibility in the lifestyle of someone for two weeks or as long as you want to run the treatment so you get an idea of whether the therapy succeeds regardless of your lifestyle or against a person's way of life.

Pharmaceutical firms are very keen to partner with us as they will theoretically collect responses quicker during a testing process and even reduce the number of topics.

We're curious about pharmaceuticals, therefore.

We also want to collaborate with employers, with coaching solutions and, finally, with payers [such as insurance companies].