covering South by Southwest 2017.

Brought to you by Intel.

Now here's John Furrier.

>> Welcome back everyone.

We're live here at the AI Lounge with Intel, #intelai.

This is the Cube, I'm John Furrier.

Our next guest is Michelle Bacharach,

who's the co-founder and CEO of FINDMINE.

retail start up out of New York City, entrepreneur.

Welcome to the Cube, thanks for joining us.

>> Thank you, thanks for having me.

>> So we're at Intel, Intel AI.

Pretty packed here, isn't it?

>> Yeah.

>> Pretty crowded.

>> I think it's the cover from the rain.

>> Yeah, it's a little rainy here, yesterday was hot.

You got a panel here later in the afternoon

about AI and retail and convergence,

but I want to ask you as an entrepreneur,

what got you into starting this company?

Was it an itch you were scratching,

was it a vision, was it something

that you felt compelled to do?

Give us the story of FINDMINE.

>> Yeah, it's actually a little embarrassing.

It kind of sounds like the most selfish reason

to start a business.

It's because I had a problem I wanted to solve,

but I think that's the best way to start a company,

honestly, because it means you're going

to be a passionate about it,

you're going to be a user of your own,

whatever you build, and for me,

that challenge was I would buy, you know,

like my silk bomber here with this big flower on it,

and I'd be like yes, I love this, this is great,

and I would get it home, but I wouldn't have

tried it on with, you know, the pants

and the shoes that go with it,

so when I'd get it home, I'd be like uh oh,

now I have to figure out how to put

an outfit together around this to wear it

and feel confident.

I think a lot of women, especially,

have this challenge where we feel pressure

to be stylish, but not everyone has

that kind of style gene where you can just see something

like this and be like oh, I know five ways to wear that.

So I struggled with that.

I struggled with that when I would buy furniture,

even when I would buy things like electronics,

like I was really looking into buying

a drone at one point.

I was like oh, that sounds cool,

I could fly a drone, I want to learn that.

I found the drone model that I thought I wanted,

but then it comes with all this stuff, right,

all of these peripherals.

They don't all plug in to the drone,

so the research involved to figure out

how to use one product in combination

with another product was way too much work,

and I figured someone should be automating that

and help a consumer like me answer the question,

how do I use this for any product

that I might pick up on the shelf.

>> And so that was the catalyst.

Where is it now today, what's the status of FINDMINE?

>> Uh yeah, that's a good question.

>> John: Solving all the problems, did it?

>> No, not yet, close.

No, but, so you know, that was like seven years ago

that I started noticing this problem

in my personal life, then I researched

and found that tons of other people have this problem,

customers will buy 170% more

if you show them how to use the product that they're buying,

but I didn't have the tools to solve it.

I have a product management background,

but I wasn't a computer scientist, a data scientist

to actually execute it, and so I'd met a friend,

a friend of mine's husband is a computer scientist,

and I sort of like, you know, suckered him in

with like this one little project,

and then he was like wow, this is really interesting.

He cares nothing about fashion, by the way.

Like he wears his Columbia sweatshirt

and jeans like every single day,

so he doesn't really feel the problem

the way I do, but what he saw was

this opportunity to use artificial intelligence

and machine learning and technology

to solve this really interesting problem

of like, can we make a machine replicate

what a human does, which is like figuring out

what's stylish, and then that's what hooked him in

and he thought the problem and the application

of the technology was so cool.

So that was, you know, in 2014 we started working on this.

Since then, we've, you know, launched a product,

we have customers on board.

We work with fashion brands and retailers.

We produce revenue, we raise money,

we have a team now, we have a real office.

We're not working out of our apartments anymore,

so it's going well.

>> So now you're in the middle of this AI world

and if you think about the data

your problem that you were originally solving

actually applies to a lot of things,

whether it's learning, healthcare,

so it's kind of like the data drives

more opportunity to collective intelligence.

Is that kind of where this is going?

Do you see that trend where it's the data

and the algorithms, or the algorithms and the data?

>> Yeah, I think that access to the data

is the big factor, so in retail there's tons

of data, right?

Transaction data, product data, user data,

all that kind of stuff,

and a lot of it is very easily accessible.

It's not all like private information,

customer information, that you have

to guard really closely.

Obviously there's some of that

because you're doing transactions,

so it's credit card information,

there's location data, you know gender,

all that kind of stuff,

but the product data is publicly available.

So we didn't even have to have

a customer live before we started doing cool stuff

with machine learning, with large data sets

because we would just go find products

that were live on the internet and use that data.

I think in different industries like healthcare

it's a lot harder to come by the data

and there's a lot more concerns around it.

>> Michelle, what are some of the learnings

that you've had, now if you look back

from where you from where you were.

What are some of the key learnings

with the venture you're building,

around what was surprising to you,

what popped out as value?

Was it the machine learning?

I mean, what were some of the learnings you can share?

>> I think in general, my best piece of advice

for start ups is just don't die.

And I say that a lot and people laugh,

but it's so true.

I've seen so many friends with startups

that kind of had a moment where they were like okay,

it's all falling apart, and they just,

they said okay that's it.

But if they had stayed around for like five more days,

10 more days, 50 more days, how their fortunes

could have changed is incredible,

and we've gone through that,

I've seen other people go through that,

so that's number one.

And the number two is, like don't wait.

Just do something.

So I think for a long time we were sort of like

waiting to get like the right data sets

in the right order and like getting it all perfect first,

and that's not the right way to approach it.

Just go.

>> So get a horse on the track and at least run the race,

get something going.

>> Michelle: Yeah, exactly.

>> And don't run out of cash.

As I always say, you can't go out of business

when there's money in the bank.

>> Michelle: Yup.

>> So, okay, so now on the tech side.

What has surprised you on some of the amazing things

that are now starting to come into visibility for you,

and what do you see as your vision?

So what's kind of obvious and that you're going after,

and what are some of the things

that you see in your vision that others might not see?

>> So what's really, what we're doing right now,

and every startup needs focus,

you can't do everything at once,

but you need to have this bigger vision

to make it, you know a billion dollar potential

kind of exit company because that's what

people want to invest in if you want

to take venture capital, and not every startup needs to.

You can self finance a business.

But for me, this rapid growth was really important,

and so I think what was really important

was that we kind of like built something

that could scale long term,

so this broad vision of like every single product

that you could pick up off the shelf

as a consumer, you know exactly how to use it.

For me, there is like a personal mission in that

because I hate waste.

I went to Berkeley, like we talked about before,

so I have a little bit of that like hippie mentality,

and I was buying all this stuff like in fast fashion,

and it just sat in my closet

and then I'd throw it out or I would never use it,

and that made me really bummed.

And the reason I was throwing it out was

because I didn't know how to use it,

and if I had just gotten that piece of information up front,

then I probably would have been able

to integrate it into my life,

and I wouldn't have thrown it out.

So doing it across all industries in retail.

>> So really efficiency too is key on this?

>> Yeah.

>> You could actually accelerate that.

>> Absolutely.

>> So on the fashion side, is that where

the focus is now on the retail side, or only still?

>> Yes, so we're B2B, we sell to fashion retailers

and brands.

They use our technology and then they figure out

where they want to get it into the consumer's hands,

so it might be on the e-commerce page,

it might be in the store, it might be

in the associate's phone, so that you as a shopper

don't even know that like a customer,

or that the associate is like kind of cheating, right?

They're looking at FINDMINE to find out

what outfits to recommend.

They might just be having an interaction with you

like a human does, but they're using

an assistive tool to get that efficiency

that you mentioned before.

>> So you have a panel coming up this afternoon.

Without giving away all the content,

what's the topic that you want to talk about?

>> So the panel is artificial intelligence for good,

and ours specifically is autonomous world,

so it's about the automation that's kind of

all around us and becoming more ubiquitous,

and how artificial intelligence is making that possible.

>> So I always get, I'm so amazed by autonomous vehicles

because I think, you know, it's so obvious,

mental models, we all have cars.

>> Michelle: Yeah.