This couple is on a mission to create helpful A.I., and they are starting in the grocery store
Nasrin Mostafazadeh and Omid Bakhshandeh are the founders of Verneek.
Category
🤖
TechTranscript
00:00 been obsessed with making human language the new machine language and we literally couldn't
00:05 let go of it and I'm so thrilled that the technology is finally working.
00:14 Our mission is to build the most helpful AI that augments human knowledge for anyone,
00:20 anywhere at any time.
00:22 What we have done is to build this AI platform that can basically make these very seamless
00:29 interfaces that you can put on top of any digital environment, any data that you have
00:34 in any shape or form and make you as a consumer to make better and faster decisions using
00:40 voice and text interfaces.
00:42 So the very first use case that we could spin off of our platform happens to be in the shopping
00:47 domain so in the commerce, so e-commerce or physical retail and brick and mortar stores.
00:52 So what Quinn Shopping AI does is as follows.
00:56 It helps you get answered to your personalized questions with the most accurate up to date
01:02 information.
01:03 For example, a mom of four can ask what is the healthiest peanut butter I can buy that
01:09 costs under $4 in this particular store and then Quinn literally immediately instantaneously
01:15 gives you that very particular peanut butter that is by health metrics the healthiest.
01:21 The questions could range from that being about product assortment to any store questions
01:27 such as customer service questions like when do you close the store, what is your return
01:31 policy?
01:32 It could be about recipes, it could be like I'm looking to cook a meal that costs under
01:38 $20 and my kids hate fish.
01:41 So something that we have found is that Quinn in the places that it has been launched right
01:46 now is most loved by store associates themselves.
01:49 They love it and this is helping them to kind of do their real job better, make the store
01:54 run better, manage the inventory better.
01:56 So bringing that gap of like knowledge and helpfulness all together it just helps everyone.
02:02 The hardest problem for AI is actually the easiest ones for human being.
02:08 So there's this thing called Moravik's paradox that tells you that language capabilities
02:12 and doing basic reasoning that a three-year-old child can do is the easiest for humans but
02:17 the hardest for machines.
02:19 So we, including myself, I switched gears working on language AI ever since and so the
02:26 rest is kind of history.
02:28 We like all of our time, like all the time that we had back in Iran we were like kind
02:32 of working extremely hard to make sure that we can like accomplish like the biggest thing
02:38 that we can accomplish.
02:39 We were always like looking for the what is the biggest opportunity and biggest opportunity
02:44 for us was the hardest problem and hardest problem just going after the hardest problem,
02:49 every time that we could.
02:51 We thought that we might be able to solve that problem that we came to US 10 years ago
02:55 to do our PhD and kind of go after our dreams and right now we are sitting here.
03:01 Say that having started a company, especially in the middle of the pandemic, has been the
03:11 hardest thing we've ever done and the most rewarding one as well at the same time.
03:17 It's just been incredibly satisfying to go through this roller coaster that it is to
03:23 basically navigate your way to start something from scratch and if you really don't love
03:30 it you will fail.
03:32 The only reason we are doing what we are doing is that we obsessively love it.
03:38 If you don't love it every day that you wake up it's going to be a tough journey.
03:43 [Music]