AI and Machine Learning
Ask about Knowledge at Wharton
Episode Description
📂 Technology
Read the full transcriptHide the full transcript
- we're speaking today with Justin got Schlick who leads the Machine programming research team at Intel labs
- the newly formed research group focuses on the promise of machine programming which is a fusion of machine learning
- formal methods programming languages compilers and computer systems Justin welcome to knowledge at wharton thank
- you so much for being with us here today wonderful thank you for having me McCool
- so I just read that description of what machine programming is but I think that you know given all the buzz around AI a
- lot of people are familiar with machine learning but most of them like me don't have a clue what machine programming
- means perhaps you could explain the difference between the two certainly yeah so at the very highest level
- machine learning can be considered within the subset of artificial intelligence and there's many different
- types of machine learning techniques one of the most prominent right now are these things called deep neural networks
- and that's a lot of what people are using to make the tremendous progress that we're seeing over the last decade
- the machine programming is really the idea that we are trying to automate the development and maintenance of software
- so the fundamental difference between the two is that with machine programming you can kind of think of all of the
- field of machine learning being a subset of the field of machine programming but
- in addition to using machine learning techniques which are these approximate types of solutions we'll also use other
- things like formal program synthesis techniques that will give us mathematically proven correct pieces of
- software and then between those two points you can kind of think of those as a spectrum you have the approximate
- solutions here and the precise solutions here and then there's a fusion of a number of different ways that you can
- combine these and every one of these things essentially is a part of the bigger landscape of machine programming
- so if I understood you right machine programming is when you create software that that can create more software right
- how would that happen and I wonder if you could give us a couple of examples just to help our audience understand
- this yeah so the idea of creating software that creates its own software is really the core idea between or
- inside of machine programming and a couple examples of that is we recently built a system using genetic algorithms
- and what it does is it'll allow you to take certain input output examples and then by running through a number of
- iterations we call them evolutions in the genic algorithm space it will then automatically synthesize the program
- that will match the input and output you do this sort of in the training phase and then it will take new input-output
- examples that it's never seen before and then generate these new particular types
- of programs so that's like one example that you might have in the space of machine programming and you would think
- about the impact that machine programming might have on different industries which industries do you think
- are likely to be affected most by this and over what period of time yeah this so this is a fantastic question and one
- that could require a very long response but I'll try to keep it slightly abbreviated at the highest level one
- could imagine that any of the industries that are predominantly based in software
- are going to benefit this benefit from this tremendously there was a recent survey that was done in earlier 2019
- that showed I think we have something like half a million computer scientist positions that are open so these are
- programming positions in industry that we need to fill but we're only producing roughly 10% of the actual programmers to
- fill those roles so what we're having in the software industry is essentially a bottleneck of supply if we can start to
- automate some of the simple tasks you know reading in a file parsing data helping us automate that
- software development testing this will I think tremendously accelerates the rate
- at which software is being developed so I would say that that's probably the first very obvious place the other area
- that I think is going to be impacted tremendously by this are autonomous systems in most of the spaces in
- autonomous systems a core ingredient of those systems is software so for example
- if one were to think about autonomous vehicles a large part of what's holding us back from getting to level 4 level 5
- autonomy which is the point where the car can essentially handle all of the nuance behaviors of driving in downtown
- Philly or something of that nature a big bottleneck of that is really the implementation and the algorithms of the
- machine learning systems if we can automatically construct those these autonomous systems will also accelerate
- in their advancement I'd like to come back in a little bit to the question about autonomous vehicles and the impact
- on the auto industry but given the fact that a lot of our audience is in the financial services industry I wonder if
- we could go a little deep into that so I mean I know that AI has made quite a significant impact for example in areas
- like fraud detection right do you think machine programming can also have a major impact there and if so what what
- might that look like absolutely this is a fantastic question and as I was mentioning earlier with machine learning
- systems their foundation is essentially learning through statistical analysis in
- some sense this makes them probabilistic that they are getting us close to the the right answer but there's in many
- cases we're not guaranteed that we'll have the right answer when we think about machine programming we go back to
- that spectrum of how we have probabilistic to very precise solutions it's my belief that for example in the
- financial sector there are certain cases where probabilistic solutions aren't sufficient so one might speculate that
- as you're doing some sort of financial transaction having the probability that oh i sort of
- been rounding off the sense right you know i'm close enough probably isn't sufficient if those transactions are
- happening billions of times a day in that case we need a more precise solution and this is one of the areas
- that i think you could use machine programming an interesting coming back to the other industry that you mentioned
- and the auto industry what kind of impact do you think the machine programming will have on the whole drive
- towards autonomy you i think you started talking about that in the context of philadelphia yeah but i wonder if you
- could go a little more deep into what that might look like for sure so as i was mentioning earlier
- we recently built a system that is using this dis genetic algorithm to automatically construct programs what i
- didn't mention is one of the pieces that's part of genetic algorithm just this thing called a fitness function the
- fitness function is essentially you can think of it as a way that you grade the accuracy of the programs or the results
- that the genetic algorithm is giving so genic algorithm produces a result and the fitness function says you get a b or
- you get an A historically though fitness functions have been written by humans and not just any human really expert
- machine learning humans oftentimes what we find is the complexity of the problem
- you're trying to solve is directly related to the complexity of the fitness function so one could naturally infer
- then why would you write the fitness function just solve the problem yourself so we took a look at this and what we
- did is we figured out a way using machine learning that we could automatically create the fitness
- function without a human involved so now going back to your question if you think
- about this type of thing in the autonomous vehicle space one of the things that's holding us back is the
- advancement of ML systems and historically the advancements that we've had with ml systems have been through
- humans creating them but if we use machine programming one could imagine like we have with the genetic algorithm
- solution that the machine can actually start to invent it's own machine learning systems that
- will then accelerate the progress of these autonomous systems so what are the implications of that well one of the
- things that I've heard about that's holding back autonomous systems as you are the autonomous vehicles is the fact
- that it might be too late to too far for the system to make a certain decision yeah and and you because you don't want
- to actually hit something right you you you you probably need software that can predict what's about to happen before it
- actually happens is that one of the issues right here absolutely right so this is tremendous insight and and
- actually we had a nurbs paper nerps is one of the leading research conferences in machine learning we had a nurbs paper
- in 2018 that tried to start to address this problem so historically what you're describing here is the space of anomaly
- detection so the Antonidas vehicle space when we think about these various behaviors we
- think this is an anomaly and in particular that's a time series anomaly so for example you're trying to prevent
- this vehicle from colliding with this other vehicle or make sure it doesn't hit a pedestrian and as you pointed out
- it's too late if you already have the event happened to you know detect it so what we did in order to address this is
- we recreated the mathematical foundation for anomaly detection specifically for time series and so with this now what
- our hope is is that the community will adopt the new mathematical foundation we've created and then they can apply
- this for time series and Omni detectors which will start to address those types of problems so and machine programming
- helps at all this absolutely so in the context of autonomous vehicles you could think of using this mathematical
- foundation to better predict these anomalies but when you think about machine programming or you think perming
- in general many of the problems we're seeing today with software is we have these correctness bugs we have security
- bugs we have these privacy violations all of these things in some sense are time-series in nature
- program is really just a sequence of instructions one after another so if you take that mathematical foundation you
- can also apply it in the space of machine programming which is exactly what we're doing well one thing I'm
- curious about is that machine programming like a lot of other areas of AI has been around since the 1950s right
- what's behind you know the sudden interest in machine programming now why is it picking up in such a big way and
- why is Intel you know so interested in sort of investing it in in it in a big way this is a fantastic question and
- I'll try to break it down to those two pieces first address why we're seeing the you know the resurgence of this
- because it has been around since the 1950s and then why is Intel so interested if we look at why it's taking
- off today I would say principally there's two reasons the first is I believe we're at an inflection point and
- the second is I believe my colleagues and I at Intel Labs and at MIT we've made an important sort of observation
- and how to think about the future of machine programming so as far as the inflection point goes we believe there's
- really or my view is really that there's three things that have created this the
- first is we have tremendous advances in algorithms in machine learning and in formal methods things that didn't exist
- say 12 months ago potentially are fundamental to the advancement of machine programming the second is we
- have tremendous advances in compute today as the recent Turing Award winners Dave Patterson and John Hennessy how are
- pointing out is we're living in sort of what they call the Golden Age of computing which they refer to as they
- call them domain-specific architectures that for a long time it was really just the CPU but now we're seeing based on
- the advances were having a machine learning in other areas we have these accelerators that are specific to these
- domains and so it's creating a tremendous opportunity for acceleration of machine learning and formal methods that
- wasn't possible and then the third piece is the abundance of big and dense data for example there is a repository that's
- called github and github basically as a place where people store their software and what we've seen by looking at github
- is back in 2008 I think it had roughly 33,000 repositories 2009 when I looked at it earlier this summer I think it was
- somewhere around over 200 million which is yeah it's it's a tremendous growth it's it's actually nearly a four order
- of magnitude growth in a decade and that kind of of growth and data as you probably know data really drives a lot
- of these machine learning systems so this is created essentially a vehicle in which we can start to explore this space
- so that's the inflection point but now goo the other point of the thing that's Intel and I think MIT observed is
- fundamentally what we're seeing is the way we've historically done programming we think is flawed that there's
- essentially a blurring of the programmers intention with these algorithms and with the system level
- details what we really want to do as we move forward is we want the programmer to just specify his or her intention
- that you you want to create a program that will tell you where the nearest Starbucks is and you just say computer
- create a program that will always notify me when I'm near Starbucks and then the
- computer handles all the details of the algorithm to implement it understands how to translate that to work on the
- hardware that's on your cell phone or in a data center that type of thing so that
- those are the two pieces that we think really are creating this opportunity for tremendous growth in machine programming
- now to your second point about why is this interesting to Intel until obviously is very interested in advances
- in hardware and me being at Intel for about a decade now one thing that I've seen which is really exciting
- is Intel really used to be just a CPU company and we're not today today the heterogeneous hardware landscape that we
- have at Intel is enormous we have neural network processors we have neuromorphic processors we have
- GPUs we have a variety of accelerators we have FPGAs and we have a ton of CPUs the problem though is programming these
- things that we can have all this tremendous harder but how can we possibly expect that the average
- developer can program this and this is really why machine programming is essential to Intel as Intel understands
- that with this new heterogeneous Hardware landscape that really is required to advance all of the
- technology that we're seeing we need a way that it's simple enough for the average programmer harness this
- massive amount of heterogeneous compute it sounds amazing the ISO since he was speaking about the work that you did
- with MIT I understand that you wrote a paper called the three pillars of machine programming mm-hm
- some time ago about some of these concepts and I was wondering if you could share some of the main insights
- from that and how they relate to some of the things we're talking about certainly
- yeah so back in I think it was 2017 a few of us from Intel labs teamed up with several people at MIT and we came up
- with this vision of what if we did this thing called machine programming what would the landscape look like and the
- main reason for this is we were seeing in these research venues that people were starting to explore machine
- programming but they were a bit disorganized that there wasn't structure around the way we were thinking and so
- the paper that we wrote the three pillars of machine programming is essentially a road map on how we want to
- express and explore the research space there are three pillars intention invention and adaptation the intention
- pillar is really what we would think of as the programmer is do in the future I don't really call these
- people programmers I call them software creators because at the end of the day our our blue sky vision is these folks
- won't write a single line of code they will express their intention either through natural language gestures visual
- diagrams whatever is best for them and and for those hardcore programmers out there they can still write all the code
- that they want so that's the intention pillar the invention pillar is it then takes the programmers or the software
- creators intention and then translates that into the actual software these are the algorithms the data structure is
- that type of thing once that's established then that work gets handoff to the adaptation pillar the adaptation
- pillar then takes that code and then figures out ok what is the software and hardware ecosystem look like for this
- particular program how do we need to augment it to make it run efficiently securely correctly and then in the
- machine learning context accurately now in addition to Intel I'm sure there are other companies that are also working on
- machine programming and I wonder if there are any companies with whom you collaborate whose work you could talk
- about just to explore how this field is evolving absolutely so yeah we have many
- collaborators in industry as well as academia and some of our industrial partners that are looking into this are
- Microsoft and Facebook so Microsoft's they have a wonderful gentleman over there submit Gulani who's seen in in
- many people's eyes is one of the founders of formal program synthesis and he actually has developed the system
- inside of Excel that will automatically figure out what the users intent is they
- call this flash fill so this is a good concrete example of real-world evidence that this is not just a research toy you
- can actually build this into real products so Microsoft is deeply interested in this another company is
- Facebook so Facebook is actually doing tremendous work in this space they recently published a paper about
- a system called aroma and what aroma does is essentially works along the same lines of the three pillars it's
- principally focused on trying to help with the intention that a programmer has an intention of trying to write some
- code and doesn't quite know exactly how to write that code the aroma system then will take a little
- bit of that code and do an analysis over a very large database and then send the
- user back is this what you meant and it's sort of a human-in-the-loop machine learning approximate solution that's
- sort of good early evidence that there's a lot of while while we think of the space machine programming as being a
- very long journey there's things that we can be doing today in industry that could be extremely valuable
- that sounds pretty remarkable again know you spoke about several companies which
- countries do you think are making progress that you find impressive in the area of machine programming in general
- in AI I've heard that China is you know advancing in leaps and bounds could you talk about you know what's happening in
- other parts of the world and what are some of the things that you are paying attention to absolutely
- so as you pointed out China is doing tremendous things one of the things that they're doing is they have a very strong
- sort of governmental infrastructural support for AI and it's my belief that the u.s. also has this but maybe not to
- the level that China does it's something that I think as a country we probably need to be a little bit more aggressive
- and progressive about there's also a lot of involvement and advances that are happening in Europe and that's also tied
- in both with their academic schools they have very strong ml leaders and academia
- and then also the vision that they have through their governmental infrastructure which European country is
- you think are doing the most interesting work that that's a great question off the top of my head you know Germany is
- actually doing some really tremendous stuff as one might imagine part of that has to do with the fact that
- they've been deeply involved in autonomous or vehicles and the natural evolution is autonomous vehicles and
- then the byproduct of that is deep engagement in AI and machine learning no D which innovations and machine
- programming do you think most promising and where do you think the next breakthroughs will occur in the
- immediate future yeah so this is this is a really fascinating question and as I was mentioning before there is a lot of
- low-hanging fruit where we can make advances and we can build things like aroma or flash fill that are very useful
- but there are some core challenges that at least with the folks that I'm interacting with that places like
- Stanford and MIT and Google and deeply within Intel labs that we don't quite have the answer to the first is the
- structural representation of intention so what I mean by this is oftentimes when we're writing code
- the programmers intention is diffused across of the code what we really want to be able to do is understand how to
- properly represent the user's intention and we don't quite have this there's a lot of advances that we've made
- historically with things like compilers and static analysis tools that create different sort of graphical or tree
- structures but when we've tried to apply these in the space of machine programming they don't quite fit that we
- can sort of you know push the you know square peg in the round hole but it's not the right match so I didn't tell
- we're thinking about this thing that we're sort of roughly calling the abstract semantic graph and the idea
- here is this structure whatever this is that we don't quite understand will be some sort of graphical representation of
- the semantics essentially the intention of what the user wants once we figure out how to build this thing my belief is
- the field of machine perming will see a tremendous spike of growth so a lot of people are working on this I'm working
- with collaborators both in industry and then in academia folks at Penn Berkeley MIT we're all
- thinking deeply about this hopefully we'll be able to figure out this this abstract semantic graph soon and until
- we do I think that we'll just sort of work with the maybe not perfect solutions and try to you know edge our
- way forward so if you figure it out what might some of the implications be so some of the implications will be the
- programs that we'll be able to generate are likely to be orders of magnitude more complex than the ones that we can
- create today for example in the space of formal program synthesis or approximate
- solutions for machine programming we may be restricted to let's say programs that
- are up to maybe a hundred instructions or less if we figure out how to build this abstract semantic graph it's my
- belief that we will move from hundreds to thousands potentially millions of lines of code that so the implications
- of this thing they're enormous so when any new technology comes along especially a machine learning or as you
- described machine programming very often technologists have to justify these investments to the CFO or to the CA CEO
- not just in terms of this is very cool technology but right this fits in with the ROI of where the business wants to
- go or order fits in with the business strategy what are some of the metrics that you think about in terms of what's
- the ROI in measuring the ROI of machine programming right this is a great question and of course as the leader of
- the machine programming research group at Intel it's my job not only to work on the research but then also justify its
- business value as you pointed out and one of the things that you you might know is Intel's very interested in
- performance but we're not just interested in Hardware performance we're also interested in software performance
- so one could imagine that if you have a programmer that's writing code that's slow
- they might blame the Intel processors is being slow even though the problem is not the processor it's actually the
- software one of the promises of machine programming and we're seeing early evidence of this is that the code we can
- generate through these automated methods will be superhuman in their performance
- correctness security so on and so forth one concrete example of that is one of my colleagues Andrew Adams
- Jonathan Regan Kelly cave on photo Holly and Diaz are folks from MIT Stanford and
- Facebook that were actually I think Andrew Adams just pivoted to Adobe research they have built a system called
- halide it's a programming language that separates out the programmers intention from the actual scheduling of that
- intention and in their recent paper this year that was published or I think in June they've shown for the first time
- that the world's foremost experts in this programming language can't compete with the machine that the machine is
- producing code now that is regularly more efficient and I think it's by I'm gonna be I'm just gonna guess here by at
- least 50% it might be upwards of you know a hundred percent faster and this is the first time in the decade that
- they've been working in halide that they've been able to achieve this but so this gives us promise that if we can do
- it in halide maybe we can generalize this and start to improve the efficiency of code everywhere this is really
- important to Intel because obviously we want everyone software to run as efficiently as possible and we don't
- want people to mistakenly believe that our hardware slow when actually the problem is somewhere else no I I I hear
- you and it actually reminds me of a broader concern that I have often heard about AI which is that as a lot of
- automation begins to take place and AI systems get implemented that the impact of jobs could be considerable so for
- example but autonomous trucks it has been a fear that lots of truck drivers could be losing their jobs if
- you know autonomous vehicle stopped hauling you know goods across the highways so do you think that there is a
- risk that if machine programming takes off that the same thing could happen to software programmer jobs and that this
- this is something that the industry should be concerned about right this is an excellent question and one that I'm
- asked quite often my my honest opinion is actually the inverse will happen is that through machine programming we will
- create many jobs perhaps millions or tens of millions of jobs and the reasoning is actually very simple right
- now we have a global population in the billions yet the programmer pool today is a very small percentage I don't know
- the exact percentage but I think that it's roughly around like 1% of the global population with machine
- programming what we are are trying to achieve is enable the entire global population to create software for
- example my mother she's incredible you know entrepreneur she's created several businesses and has just done fantastic
- but she's not a programmer so the entire world of software is closed off to her and in fact this is one of the reasons
- that I became so compelled is I see someone like her who's wildly creative she has some amazing ideas but because
- software's closed off those ideas never get realized and hopefully with machine programming with this intentionality
- that we're discussing earlier this will this will create tens hundreds of millions of jobs it'll also keep the
- programmers that we have today employed because there is work to do on building these very complex systems and as we
- expand intentionality we're going to require those people those what we call it Intel those ninjas to be ensuring all
- the subsystems that are part of those three pillars are advancing appropriately so since we were talking
- about these advice consequences it reminded me of a conversation almost 15 years ago with
- Andy Grove the former CEO of Intel whom we had interviewed in 2004 and and he once said that for every metric there
- should be another paired metric that addresses the adverse consequences of the first so as you were thinking about
- some of the metrics that you would use to measure the success or the ROI yeah off of machine programming what would be
- some of the adverse consequences of machine programming and what metrics would you use just to keep an eye to
- make sure things don't get out of control I'm really glad you asked this question and on top of that before I
- answer I just want to say I'm a huge fan of Andy Grove it's wonderful to be at a
- company with such strong legacy of leadership and you know we see the impact that Andy has had even today that
- the company is really trying to follow a lot of his principles I would agree with
- you wholeheartedly and in fact in a book that we wrote in 2004 that knowledge at
- Wharton wrote he was identified as the top leader among 25 leaders in the past 25 years oh that's fantastic
- yeah I never got the chance to work with him personally but I know people that have and from everything I hear he was
- not only a tremendous leader but also tremendous technologist which is sort of a very rare combination but going back
- to your question about the adverse consequences we actually talked about this a little bit in our three pillars
- paper and this is part of the reason why we we wrote this paper is that what we were seeing in some cases of advances is
- there would be research for example one of my colleagues that I'm very fond of Alvin Chung he's a professor at Berkeley
- and he's doing this work called verified lifting and verified lifting essentially
- uses formal program synthesis techniques to lift code from one programming language and then drop it down into
- another programming language this is wildly useful for legacy systems that can't be maintained because we don't
- have a programmer supply we can lift that code out put it in language where we have lots of
- programmers however one of the things we noticed and I've discussed this with Alvin so he will probably not be shocked
- when I mention this is that there's a potential byproduct of that lifting that can reduce intentionality so for example
- his work we would say principally falls in you know the invention and then the adaptation based on how that code is
- transformed the intentionality of that code could be reduced for example things like variable names function names
- things that are really important to programmers may not map properly to the new structure so as we're making forward
- progress in machine programming what we've asked the community to do is think of the context of the three pillars and
- then try to understand are you inadvertently hurting another pillar and if you are clarify that so we understand
- that this is another thing that we now need to advance now since you mentioned you know some of the academic
- collaborations I was I know that you spoke at Penn that did precise event last week and I was wondering if we
- could end by talking a little bit about what kind of work you are planning to do
- here at Penn oh yeah so I'm yeah very delighted that you asked this question and I'm really happy to be here
- I had the honor of giving this talk at the precise industry day which was over sold out I think people were sitting on
- the floor it was just very well attended and rightfully so the lot of the thought
- leaders in the space of computer science and formal methods in machine learning are part of the precise center recently
- I accepted invitation to help chair the technologists regroup for precise and also act as their executive director for
- artificial intelligence my role with precise and with Penn actually I think is twofold the first is with precise
- they have a very strong sort of technical Horsham of industry collaborators and what I would like to do is ensure that
- all of the industrial partners are working in a very complementary way that we understand what the core challenges
- are and that we're not working in a way that's overlapping duplicating effort so
- that's one part the other part that's really important to me is right now we have sort of a lack of machine
- programming engineers and researchers in fact there's very few of us which makes
- sense because the field even though it's been around since the 50s it's had struggles to get to the point where it
- is today so what we're working on with Penn and other academic Institute's is to start
- to incorporate curriculum changes and get our undergrads get our grad students more familiar with it and then also
- generate the new leading minds through the PhD programs that are then going to drive the research they're happening
- both in academia and in the industrial labs that sounds wonderful and Justin thank you so much for taking the time to
- explain all these things to us it's wonderful to meet you and very happy to have you here at knowledge at Wharton
- yeah it's been my pleasure thank you so much for having me for more insight from
- knowledge at Wharton please visit knowledge Wharton UPenn edu [Music]
Brief Summary
Justin Gottschlich discusses machine programming's role in automating software development and its implications for various industries, including finance and autonomous vehicles.
Key points
- Understanding Machine Programming. Justin explains the difference between machine programming and machine learning.
- Impact on Industries. Machine programming could revolutionize industries reliant on software development.
- The Three Pillars of Machine Programming. A roadmap for machine programming focusing on intention, invention, and adaptation.
- China's AI Advancements. China is making tremendous strides in AI, supported by strong governmental infrastructure.
- The Future of Machine Programming. The development of an abstract semantic graph could revolutionize machine programming.
- Job Creation Through Automation. Machine programming may create millions of new jobs rather than eliminate them.
Episode highlights
- Understanding Machine ProgrammingJustin explains the difference between machine programming and machine learning.“Machine learning is a subset of machine programming.”1:22Nov 22, 2019
- Impact on IndustriesMachine programming could revolutionize industries reliant on software development.“Any industry based in software will benefit tremendously.”3:46Nov 22, 2019
- The Three Pillars of Machine ProgrammingA roadmap for machine programming focusing on intention, invention, and adaptation.“The intention pillar is about the programmer's role in the future.”17:37Nov 22, 2019
- China's AI AdvancementsChina is making tremendous strides in AI, supported by strong governmental infrastructure.“China is doing tremendous things in AI.”21:14Nov 22, 2019
- The Future of Machine ProgrammingThe development of an abstract semantic graph could revolutionize machine programming.“The implications of this thing are enormous.”24:14Nov 22, 2019
- Job Creation Through AutomationMachine programming may create millions of new jobs rather than eliminate them.“Machine programming will create many jobs, perhaps millions.”29:31Nov 22, 2019
- Curriculum Changes at PennEfforts are underway to incorporate machine programming into academic curricula.“We’re working on curriculum changes to get students more familiar with it.”36:12Nov 22, 2019
- Closing RemarksA warm conclusion to the insightful discussion on AI and machine programming.“It's wonderful to meet you and very happy to have you here.”36:41Nov 22, 2019
Tension
45/ 100Some tension





