Transcript
Kyle:
[0:22] Depends on your language.
Cory:
[0:33] Sitting incorrectly I’m not I don’t know why it’s not.
[0:41] Sivan and what’s the last name Metzger Savannah Metzger and Nisha and say your last name.
I’ll just dinner dishes Savannah Niche if it’s okay with you I’m at the Sparks.
Sivan:
[0:58] I’d over here last year.
Cory:
[0:59] Where you left okay was it here at this location last year.
It was okay very cool so this is Corbin from the big get a beer team we are at spark Summit spark and a ice I’m going to keep messing they apart up I’m joined by my friend, Prince,
hey there brother and we are excited to have two Executives from a very interesting company called parallel M and Savannah is CEO and co-founder and Nisha is CTL and co-founder welcome to the show.
Sivan:
[1:26] Thank you my dear.
Cory:
[1:27] Excellent so tell me a little bit about yourself where you from what are you how did you end up as the founders of a tech company.
Sivan:
[1:35] Well I from Israel and.
You know I’ve been out working at a lot of sun in southwest and Center 2014 large companies medium companies in at some point came in time to work in a smaller company in take everything on for responsibility.
Cory:
[1:52] That’s one so where’s home for you here in San Fran.
Sivan:
[1:55] I love you for many years but actually we moved back to Israel couple years ago I landed here this morning.
Cory:
[2:00] How does Yokai x 100.
We can tell you’re doing great that’s incredible you look much better than I do after a long flight like that look like a truck ran over me how about yourself.
Nisha:
[2:11] I’m so I basically I mean I’ve been in kind of the software space so I actually moved out here to the Bay Area to study at Berkeley and kind of no never and I laughed
and I’ve done various kinds of software distributed systems large companies small companies this is my small company rotation.
Cory:
[2:27] Absolutely.
Nisha:
[2:27] Well and I’m based here I live in the South Bay.
Cory:
[2:30] Excellent so you walking around if you attending the sessions half the company seem to be nurtured been some way shape or form by Berkeley grad so that’s that’s a pretty impressive organization and got there so
parallel and what is the what is the main offering the parallel lines bring it to Market and and why is it so critical.
Sivan:
[2:48] So Perla them came to be because we believe that there is a big amphetamine in the market.
[2:55] Are there many many projects many data scientist very very hibiscus expectations to deliver some kind of competitive change operation of History fish in sea based on the usage of machine learning however,
only a sliver of those actually end up making it into production,
literally and it’s and it’s not going anywhere near Pace to satisfy the need of expectation in my previous company I was a busy executive and we have 13 data scientist,
and they would come to me and say I we finished everything and I was here but nothing to benefit here,
and I think this is similar to most of the customers that we see you today and then we
Nisha and I worked a lot together to identify that the big impediment we believe is part of party decor mainly because there’s no best practice in the industry to move from academic steak and I,
to move from the data science site the actual production and delivering an actual value and this entails many things some of them would be collaboration between the constituencies in the company,
when you have data scientist you cannot and should not be expecting them to one of those really expensive resources in the company to be actually delivering 24/7 operation services,
there’s there’s other teams for that that should be brought to the table and work together and there were the software evolved over time to come to a point with his devops which now is clear for everybody so weird
coining the term ml Ops Emma lobsters sense of expenses be a category of a kind of Bridge to Bringing machine learning,
into a production using collaboration between the teams SML Ops.
Cory:
[4:24] Okay so collaboration between team so we think about the the data science and the Machine learning process right if you have lots of tools that are in the data collection and prep space and then you have the tools that are good for
analyzing the Frameworks for manicure for strictly running some query against the data and any of these there’s this this
space look like ml development tools were people use your how do you develop the algorithms it still it sounds like you’re you’re going to the next stage which is I need to collaborate on that kind of that are Lie part 2 the cycle,
but,
when you say you’re there not getting into production is because it’s hard to move a model into production or is it that it’s not it’s just not germane to the people who built the model that ain’t even know how to go get it in production.
Sivan:
[5:08] I guess it’s a combination of everything you said in your right where we looking when we started this what we look behind Beyond The Curve as in most people most of our customers that we talked to they really think she really,
first identified the problem then I’ll find it they decide then I’ll hire data scientist then I’ll hire did engineer and recently there’s new roles there’s a citizen the engineer now I heard of a data Wrangler,
text me what kind of cowboy.
Cory:
[5:30] I can’t wait to see the love the stickers and we get for dinner Wranglers.
Sivan:
[5:33] Exactly him and and we think that that’s all good but when you when you really win the rubber needs to eventually hit the road these people actually don’t know what it means to,
take into consideration resource constraints Network you know timing distribution security governance all those things which which are mandatory when you’re running an actual business
are these guys it’s not that they’re not smart enough there this morning people in the room but they’ve never,
had to or never wanted to take care of this. As a company you should not expect your data scientist to be running or operation that that should be where you want.
And the infamous from what I meant when I said around the curve is we we saw this coming,
I’m too forward the sewer progression of the thinking of our customers is going to this point and even the Delta between last year and this year here,
Lassie when you’re talking about it nobody knew what operation ization actually meant and now it’s like fun center and people are coming to us and and talk to us about it all the time.
Cory:
[6:27] Yes we’re because a lot of times when I was sending some of these shows and even we talked a lot of the vendors that are helping in that,
that tool kit area for data scientist where they’re helping to develop the models that it almost seems like the conversations for deployment always stop at yeah but we we publish a python script or we have a restful API is that is that the lights do you pick up from there and and you run those
do you think your platform run the the API framework better is it a governance platform I’m interested technically like where the handoff is aware that collaboration is with the tools for the data science teams.
Nisha:
[6:59] Sure so so basically kind of we come in you know after an organization has done experimentation,
with algorithms and so you think of the focus I mean you should people start with a pile of data right and then they believe that data has some value within it the experiment and they get some initial evidence,
but there’s inside and they know how to extract inside so I’m a lot of the tools that you’re talking about are they playing that space
there and on that’s where they add value we come in when you have you know something that you believe is in like you to generate inside and you wanted to first of all Impact your production in a business and your and also do that,
continuously twin over a lifetime of your production business so that’s why I said we pick up
you know algorithms we pick up models that have been trained by tools of such sometime scripts and so forth but and we allow them to in lb are in Urdu production in France and so forth but we also allow you to retrain,
and such in production because one time training doesn’t really work in most of the scenarios,
days of keeps changing you need to make your models better your competitors have found a better model,
you need to do better than them now in or you got more data now than you had two years ago when the first experiments occurred so it doesn’t stick like thing,
the you know the experimentation is something that you know we come in right after you’ve done some but after that we manage the whole time.
Cory:
[8:20] So the data scientist team they they start to use parallel them to like you said once I’ve gotten something useful that get a Model A straight to the point
how did have somebody deploy parallelism in that context is it something where you’re you’re taking that script and it’s running on Parallel them in the cloud is something they run internally does it what does it mean to run parallel import organization.
Nisha:
[8:41] So so basically personal we are software and we’re out softer than replacing the control pad that we provide a runtime,
production but we don’t actually run the models we allow you to run the models in any environment that you have and that could be anything from a stand-alone Python program,
running it on spark plugs and this is a practical necessity is because the space is evolving to be heterogeneous,
and fun and one of the techs virginity is actually one of his friends is that there are so many algorithms out there is no different classes of machine learning have different engines that are good for them,
and what we found a customer’s is there always looking to solve a problem they’re not really looking to standardize on one set of software so you want them to be able to run on whatever they need today,
and then whatever they need tomorrow so what we do is we are runtime interface with these engines and we manage those engines in production so you can run your program.
Cory:
[9:33] Excellent so so what did people do before parallam was there was it,
was it as you said before as data scientist or just hoping that it got into production where there are there other tools that were maybe less or maybe open-source he’s at work cutting the,
you’re cutting it yet for Enterprises what what tools have you are you seeing the parallels replacing or is it just a completely new category.
Sivan:
[9:55] What we’re seeing a first you know we didn’t invent the world,
companies like Google and Amazon having deploy machine learning production having said that now that is becoming mandatory for all the any advance company to be to be in getting involved at this we literally still today we see people that are using a week like toothpaste bubblegum in Excel,
I need the schedule there bi-weekly training of their moms and their Outlook calendars using Google Sheets to manager models that are running or not this is completely not scalable but I think the point is that.
And I think the sweetest spot is the today you have 50 or 500 they decide this working organization,
it’s amazing how many companies were seeing in the last couple of months alone that are only getting to the point where they have two or three services that have been deployed in production.
And if you would forget from a business standpoint expectation after paying so many salaries so many Technologies from the business side to see some results,
as soon as the business sense is that they are there two or three services production we here consistently we want a hundred Services by the end of the year,
and those people that attend those people that can like a deer in the headlights of thing I’ll hang on the way we did these two two things bubble gum Excel is not going to be the way we’re fixing to take us to 108 turn around and looking and,
today we are on you know one of the one of the shops in town that the only ones that are that are really focus specifically on this problem scaling your machine learning service in production.
Cory:
[11:21] So so true because when you when you talk about that,
conversation that happens if the customer year you’re bringing together two groups as you said before data science groups in the operation seems that maybe work work talking very well out front if they did maybe they have a better process is that challenge just from your from your
from your engagement with customers to actually get those teams working together and selecting the platform that that everybody can be happy with.
Sivan:
[11:45] It’s a good question and the thing is if you go if we go to Pelee only today the scientist,
they’re not really sure what to do with it if you go only two operations who have not yet been brought into the party
then I should I need to get another system these guys usually 400% over what they can do anyway it’s your thing when you get to the person who owns this like I either the Evangelist to head of digital transformation,
or we need solution architect needs to look at things 6 to 12 months of the future how they’re supposed to look they come back and say oh my God we have to have something like this,
and they bring those people around the table in and something we launched a large Bank up in,
Canada a couple months ago and it was great to see how we can do a workshop and it and they brought it and the architect and the other data science governance people business analyst all them for a whole day in one room,
and an after we kind of went to the workshop it’s either we realized they really never sat around and had this discussion before together as a group,
add to tackle this and it was the Evangelist that brought them together and and then they start asking each other and ask questions of the subject matter expert having done this multiple times in,
and we actually it was great to see how we can support all their questions and really given that understanding and I think the best compliment we came out of from that day was it oh you’re not imagining suffer you’re the way to do this,
as in the way to do production machine learning which is great compliment as ice.
Cory:
[13:06] That’s very cool so I love that the bringing together of the teams it’s it’s interesting to me even in large Enterprises and some midsize and small how little of that cross team communication actually exist in in real life
you got to have an in the space in this exciting Big Data machine learning space there’s gotta be just a ton of Integrations and Partnerships that are important for you in order for your platform to be successful what are some of the most like strategic Partnerships you how to develop or
strategic integration that have been really important to your success.
Nisha:
[13:36] So I can that I can definitely speak to the integration so I think in a one of the points is kind of really about how ml Alps
works with that Bob’s right cuz you know when we say I’m allowed to wear referring to the specific challenges of bringing machine learning into production and we are not really looking to replace
customers devops,
how to oil change and so there are a number of places where we are we being very specific to machine learning in add things but they’re also places where we caught in a collaborate very much with existing in a software life-size.
That’s really what the customers trying to put into practice so good example of integration is with get and get help customers keep cold there,
and so we walk you know completely seamlessly with that another good example is obviously the Integrations with these engines we are not looking to in our build and analytic engine to compete with spark,
we know we want to make sure that anything that anybody finds that’s running on spark can be put into production using us.
Cory:
[14:29] Libraries make it easy.
Nisha:
[14:30] Libraries go for it if you tweak them that’s fine if you brought your own from scratch because that’s that’s important because you know that one of the reasons why the space is maturing so fast is because of that innovation,
and customers want it be no benefit from that so that’s another big area or third one is essentially the the kind of the infrastructure of the Datacenter you know the security infrastructure the authentication the resource management,
all of witches you know people don’t want to change the way they run their data centers just because they want their running machine learning they just want to put the machine learning into production and have it be a good citizen,
within their practices so those are some of the areas where we went to grade it so.
Cory:
[15:08] Just what do you talk about
operation Sweet we were to the talk earlier kubernetes is clearly a as an open source technology and then as it’s being packaged and commercialized in a variety of places it seems like from a deployment perspective something that would have to be critical to your team.
Nisha:
[15:24] Container based apartments are very you know.
Cory:
[15:26] Are they is it I mean I’ve been in the biggest race for about five six years and I’ve watched it where we went from this,
you know early stage can a big data and went when we just went organizations really were just playing around with Hadoop was just like let’s Tinker with it it was always this we need no orchestration and needs to be bare metal
my commodity servers and I feel like we’ve done this whole switch the other direction now to work like I have to talk to her about how we containerize things that seems like an interesting model before you guys,
that has to complicate things cuz if now if everything’s in containers it’s more things that I have to orchestrate it’s more microservices is that is it challenging or is it does that represent opportunity for you.
Nisha:
[16:04] So I think that you are absolutely right that’s the reality right that’s the way of the world is going on I think it’s actually actually makes the machine learning deployments more streamlined,
does one thing is that for as I can and when we’d example is if you decided to build your machine learning prediction as one tiny little piece of code in the bowels of an enormous app it’s impossible to manager,
is it possible for anyone else to manage it but if it’s a microservice then it’s awhile identified thing,
can be upgraded separately can be mounted managed separately so all of those structural Trends actually make the MLS problem technically more feasible,
also enable Solutions like ours to come in and say we focus really on solving that problem and we’re not trying to manage your parent business logic.
And all about you clean Lionel model writes what you’re doing so that you know your machine learning can be managed I think it’s a win-win actually.
Cory:
[16:54] Excellent,
secure is it from a form of Eli, what’s next for you guys what you what you and the team are looking at both technically in from a from a go-to-market and Company perspective what are some of the trends that maybe you’ve heard from the show or that you’re just watching,
you know people in and around this space around machine learning and operationalizing said machine learning capabilities the macro trends that you guys are really thinking or interest in the next 12 to 24 months.
Sivan:
[17:18] Show me looking at the name of the show right and just be the spikes on it now it’s Park Nai so everyone is pushing forward and trying to move up the stack,
from our perspective 2 3 5 years ago as you mention the big thing was a scarcity of the data scientist in the biggest problem was the date of the date is so dirty it’s not that all these are gone away but they,
by order of magnitude become much better right that the platforms at the college he’s in the time of the lapsed have brought the industry further and further
and I’m to the point about us looking around around the curve,
it’s it’s really fun to see that now it’s the value level of the value surfacing in the value chain is coming to the point where I actually the spark meet the AI and actually
deliver some values of the business and that’s that’s what we’re excited about and what we were looking forward to.
Cory:
[18:04] Yeah I was.
Nisha:
[18:06] Anything that makes sense I think it’s really right now that the problem of MLS is multifaceted because machine learning is a very different kind of app,
and so I think these last you know last year we worked a lot of the nuts and bolts and so forth and we know we’re continuing to add into specifically understanding those applications and making it easier to diagnose them to understand how their behavior in which is an art form on its own,
so technically that’s what we’re focusing on.
Cory:
[18:31] Yes I am curious though cuz I know they say eyes in the in the title but is because I think I ass here have we I mean are we really doing II and it’s in the form that that I guess maybe the naysayers are afraid we’re doing it yet or do you think we’re just we’re getting there.
Sivan:
[18:47] I heard Michael Jordan this morning saying that don’t don’t kid yourself and nothing they were doing today I would only the beginning of Nino machine learning plus oh,
I think we both subscribe more to that and that’s what we called it ml Ops and not a i and in our material and everything we talk about machine learning because you want to be in with Robert actually hits the road we as an industry are deploying machine learning and so I can see the value of that,
I think it’ll evolve into whatever the definition of the I actually ends up being but,
I think it was exciting for me I think Elsa Furniture is to Prevail this industry forward to take it really to the next level to start seeing some Valley from all this talk.
Cory:
[19:23] Yeah we think it’s like yesterday and the Keynotes was that they are they put up the metric they said like less than 1% of the fortune 200 or Fortune 2000 companies are actually,
recognizing real measurable value from machine learning and AI today and when they put up the list the thing that was comical which I think separates this conference from many others was all the names up there like the names of all the companies who are actually here
representing Apple an Uber Google and all this Facebook companies so
obviously those organizations are interesting in there I had in there may be developing some of their own butt,
what’s the best like the best next step for Enterprises who are struggling with trying to achieve value for machine learning at a macro level like what are you advising obviously middle Ops can help,
you’re trying to help Enterprise understand how to move from ideation stage how does it develop a plan to actually implementing these thing.
Sivan:
[20:19] I think you know this is really what we seen continuously people thinking very seriously,
oh we don’t leave very early on we decided not to be a vertical solution people were pushing us to do a vertical solution for Banking and very early on we decided that,
if we filter the Verizon is social maturity going to be horizontal,
as in we want to buy something across the industry to help the operation like that so again people are thinking seriously the more people start thinking more a little bit more parallel or where is going down and the quicker they will get,
somewhere,
in terms of understanding value the quickly be a babysitter 8 fail correct correct understand what’s right for their business I don’t think we have the answer to the businesses,
do we have the answer to Hey try it try start seeing what happens in your computers are doing is so you better do with you.
Cory:
[21:10] Digital transformation is often driven by fear of what the not only the traditional competitors but the,
yeah exactly the fear is a great motivator to see where parallel I’m goes what’s what’s next for you guys on the road map in terms of development what are you really focused on developing against for your product of your offering next.
Nisha:
[21:29] Sure so I’m so thinking of right now you know I’m at Savannah said you know we you know we’ve creators in the first version of our product we you know we’re working with,
a lot of customers do you know doing no in terms of understanding what they need fundamental even the space as you mentioned is very is still fairly early,
what we are really focused on is making sure that we can create a good production life cycle for really the most solid basic algorithms that are the ones that are going to be in production that’s one of the extended examples of where the AI is here or not,
in all depending on who you ask about the eyes either here or not care if it’s basic solid stuff that’s in a very like you to work it is here,
if it is you know something very very fancy no it’s not here yet so the space will naturally expand places we’re looking at are really the foundational,
problems that have to be solved when you you know in terms of and areas of Diagnostics and governance and so forth where the problem is real but the existing ways to do it for standard application Stone Cold,
so I’ve got simple example is in machine learning program can misbehave without showing any outward signs,
of misbehavior the process can be up and running no issues Incipio Atlas Asian no errors but it’s predictions are bad so that’s a very soon oh it’s so simple to say it very difficult to solve,
those are the kinds of problems that are going to materialize and those are the things we work.
Cory:
[22:48] Yeah I actually found that the security talk that Don song that yesterday talk about some of the ways that people are maliciously attacking models I think it’s just a it’s a space that I honestly I’m not that bad of a guy so I didn’t thought about that but when she started bringing that to the surface,
I mean it really like melted freaking thinking about how sensitive it is it has we automate things right which is you know a I at this point is a lot about automation I think it in my opinion but,
as we automate so much we have to be at want mindful of what we have to be virgin and we have to be retraining with the understanding so I can imagine.
Sivan:
[23:20] Fuses in the system rights as if something starts if there’s a negative got to be able to either stop it or rollback or any any corrective action to avoid risk.
Nisha:
[23:28] I also understand what happened so you know so like the city of New York for example recently instituted committee to understand algorithm fairness and so that’s an interesting one because it shows so Ordinary People are worried,
the under and then I worry about algorithms touching lives at worry about algorithm that I locate budgets and I’ll grab them said decide who’s going to get promoted I was going to get parole and stuff like that and so there’s you know these are.
Cory:
[23:52] Don’t worry about the biases.
Nisha:
[23:53] That word about the buses so these are real problems that have to be no be counter before I can really grow these are the kinds of problems.
Cory:
[24:01] That’s awesome I want to ask you just a side note question before you move move on,
start of the company based in the bay living in Israel started a global company what’s that like to be an executive in a in a Silicon Valley start a buck what’s your favorite thing about,
getting up every morning and get being able to build your own business in this business climate.
Sivan:
[24:24] I think you have the possibility no opportunity is endless it’s it’s so much a matter of execution and you know from here on flight to Toronto then flying to London we’re trying to expand into additional markets,
how do you balance and juggle all that together and having the right product for the right Market attacking it timing it’s just take a chess game on you know 17 different boards the same time so.
It’s exciting at pepper in the jet lag and everything is so fun.
Cory:
[24:54] Jetlaggin espresso field conversation right I want you what what’s your favorite thing about your your your current opportunity.
Nisha:
[24:59] So so one of the things I like about startups and then I think this is true of an everywhere not just in the valley is that you can solve,
difficult new problems but do it in a very fast running,
kind of way like if you are in research way and I came from a classical research you know that’s what I know so I got my doctorate I did the usual thing and so if you’re doing research or solving problems that are hard but,
you’re the answer will take years to reach,
the world and if you are doing traditional development you basically are solving immediate problems but they’re not usually the hardest problems out there
so one of the nice things about doing a startup in particular you feel like this is you are solving real cutting-edge problems for which there are no solution,
what you’re doing it in a way that if you solve it you can read someone tomorrow that’s what I like about it.
Cory:
[25:44] Very cool we appreciate you spending time with us the folks that are interested in learning more about parallel and I’m guessing company websites probably a great place to start for demos and that kind of stuff.
Sivan:
[25:53] Pearland.com.
Cory:
[25:54] Okay and then are you guys attending any other big data shows here in the next u.s. call through six months that are that are interesting where people can find out more.
Sivan:
[26:02] When shows all the time like I said I’m in AI Summit of Toronto next week and II Summit London the following weekend when all the pictures.
Cory:
[26:10] Awesome very cool.
Sivan:
[26:12] We have webinars coming up as well.
Cory:
[26:13] Play okay that’s all put some links to those in our show notes that’s perfect we appreciate spending time with his parallel him sounds like a great platform for really marrying how do we take machine learning and,
operationalize it in the professional enterprise-class way and I think obviously solving some very interesting problems for those organizations looking to really finally achieve value,
Pro machine learning I want to shift gears in a little fun we want to learn from you especially his folks were traveling the world and and have an interesting perspective,
so we’re going to ask you some questions we called the rapid fire we ask all of our guests the same questions and the way this works as we want you to,
don’t stew over the answer like just give us the first kind of thing that comes to mind when I ask you a question cows going to
I asked a question to ask you tell me what kind of play ping-pong will go back and forth with who gets asked first.
Kyle:
[26:59] You ready let’s do this right what year will Skynet go online Skynet from Terminator.
Cory:
[27:10] Machine start taking over they start building machines to kill us never 2015 all right.
Nisha:
[27:19] Are you sure it’s not already there.
Cory:
[27:21] Simon that’s cool I thought that’s where I’m at.
Kyle:
[27:23] Yes me as well what is a great last book that you read.
Sivan:
[27:28] The hard thing about hard things Bend Oregon.
Nisha:
[27:33] Persepolis rising from the expanse.
Kyle:
[27:36] Okay awesome what genre of music are you currently rocking out on.
Coldplay Coldplay.
Sivan:
[27:54] I’m not that amazing feeling when I run to.
Nisha:
[27:59] I have a nine-year-old so Miley Cyrus.
Kyle:
[28:01] Oh yeah yeah yeah.
Cory:
[28:01] 49 year old man who turn into a little lady that stuff.
Kyle:
[28:08] What piece of technology is currently making your life worse worse so what one is behaving against you.
[28:24] Everything is functioning well so far.
Cory:
[28:26] He’s lucky one of the lucky.
Nisha:
[28:28] Alexa and Siri in my kitchen at the same time.
Cory:
[28:30] Have you got them talking to each other yet.
Nisha:
[28:34] Yes and I clearly have way too much time on my hands.
Cory:
[28:35] That’s awesome.
Kyle:
[28:38] Wow all right what is your biggest Money Pit right now.
Cory:
[28:43] Personal money that we recognize running the company is probably the Money Pit.
Sivan:
[28:48] I have four daughters pictures pic you know whatever you want.
Kyle:
[28:52] Literally anything.
Cory:
[28:53] Everything having for $4 I get that.
Kyle:
[28:55] Yes.
Nisha:
[28:56] I don’t know but I’m sure it’s related to my daughter.
Cory:
[28:58] Kids are the overarching number one answer the way yeah that’s.
Kyle:
[29:02] Yeah that’s a number one most popular answer.
Cory:
[29:05] Use that as advice buddy.
Kyle:
[29:07] We’re enjoying the,
dual income no kid life alright obviously you’re traveling here in the next little bit are you taking any personal vacations to anywhere interesting or cool.
Cory:
[29:24] Nothing will work for the man.
Kyle:
[29:25] Okay.
Cory:
[29:26] The most interesting place to travel in the near future for work.
Sivan:
[29:30] I’m here Israel London Canada all over the US nothing to.
Cory:
[29:37] Nothing too exotic.
Kyle:
[29:38] Feel like a better answer for him would be what boring city are you going to United.
Cory:
[29:43] At least interesting City Cleveland Ohio.
Nisha:
[29:47] So I’m going to rain next week.
Kyle:
[29:48] Oh wow I just came back from there. My wife and I did two weeks in Italy and when I flew back home Friday evening and then Monday morning flew back out to San Francisco so it’s.
Nashville Tennessee so I’m working through a little bit jet-lagged I understand your pain.
What TV show are you currently watching.
Sivan:
[30:12] I watched on designated Survivor.
Kyle:
[30:14] That we got sucked into that one for Susan that’s like the continuation of 24.
Cory:
[30:20] That’s what I thought it smelled like whenever I saw the ads for them I like 24 so we’ve got it on the DL DVR ready to queued up ready to go.
Nisha:
[30:28] Direction kind of between TV shows I exhausted Agents of Shield and now I’m looking for something new.
Cory:
[30:34] I will say we plugged in many times I’m a big fan of Westworld that show especially in contacts is just one of those that just makes you bring hurt and feel awesome simultaneously.
Kyle:
[30:44] Yeah yeah that’s a that one in stranger things seems to pop up.
Cory:
[30:47] Yeah I was ever thank you so much for spending time Savannah an issue was great to chat with you is there a way where besides the website where would we find you guys and gals on on the social media are you on Twitter on linked-in what’s your favorite platform.
LinkedIn.
Nisha:
[31:02] I’m also on LinkedIn.
Cory:
[31:03] Excellent thanks so much for spending time with a safe travels home and enjoy the rest of the spark and AI Summit.
Sivan:
[31:10] Thank you so much thanks.
Be the first to write a comment.