An Interview with Ishan Anand (CTO) and Mark Brocado (VP of Engineering) from Moovweb

what is up welcome back to the channel yet another video but this one is the first one for 2021 i know it's like the 23rd of january so i've missed two weeks you know what sometimes you just need to take a break but i'm back so in this video i got to interview the cto and the vp of engineering of move web which is a super cool startup so the cto ichian and the vp of engineering mark go in depth about you know what is move web why do a startup you know what type of people are they inside the startup how do they do their stuff but also what is in it for you are the audience because move web you can put them next to like netlify and for sale but they're a more scoped organization they kind of have clients that are huge e-commerce and then you still want jam stack but sometimes you also want dynamic stuff and that's what move web does really really well so without going much deeper um let's just dive in have fun with this first video of 2021 and you know i'll see you next time cheers so welcome guys to our video i'm super happy to have you here and actually my first question that i always ask is how do you pronounce your name because i've most likely not said it correctly so i think it's ishan right so let's start with you yeah it's think of the eye as a long e and it's ishan uh last name is anand uh i'm the cto and co-founder at move web yeah go for it mark mark brocado vp of engineering and move web awesome and does that sound like american italian or something am i right here the source of the name is sicilian but my parents immigrated to america turn of the century there you go and are you now still there or did because i think you're in my time zone right which is c-18 in estonia born in middle of nowhere new jersey and now i live in tallahan estonia so wow well i i always love this this part of the interview which is always the first bit because now that it's all virtual like this we can talk to about anyone like i've had people from africa on and it's it's good fun anyways so you guys are i think my second startup founders kind of or vp of engineering people i wanted to talk to and what i'm always interested in is like what made you feel like you wanted to do this thing and start a startup and go down this crazy path of do i get funding or do i just make a product people buy and then fund it that way or and like how did you guys get to start this um this this enterprise of yours yeah so my personal journey uh has been at least technically always kind of slowly working up the stack from kind of device drivers right out of school to to cloud web and and mobile here at move web um and when the iphone came out i was actually doing a self-funded uh startup trying to build the app store for the iphone before there was an app store really there was there was no app store i don't remember not having an app store yeah so uh people don't really remember this but the first year the iphone was released there was no app store there were no native apps at all there was no sdk and what jobs had said was build web apps um and uh well that's what we tried to do and so we tried to collect a bunch of them in in like an app store format and within about six months they they said mia culpa on that and and said we're gonna have a native sdk we're gonna have to wait about a year before it comes out so at that point we we folded because you know if the platform starts building the same thing you're you're trying to do you yeah this is kind of the way this goes right exactly yeah you get sherlock so yeah um so then i met up with aj the ceo uh and co-founder of move web and he was working on bringing the web to mobile um in some novel new ways and we spent uh you know we were in the app store on day one when it finally did come out and we had a product that we released helping companies go mobile in the pre-responsive web days and that did really well and then you know we've always been you know developer centric and developer first and we saw what was happening in the front-end space and we actually even saw you know for me personally we finally were getting you know pwa technologies we finally had you know here we are like 10 years later we're finally getting the the ability to build web apps the way jobs actually told us to do for the iphone i can imagine like we really wanted those features i've been there for like 15 years working on this kind of same stuff right and we never had the stuff and it's slowly starting to get there now yeah your current startup is different right well that's what got us into really the space looking at pwa technologies and and really that's the gateway into kind of headless and so you know we have a long track record of helping you know large e-commerce companies and when we saw was happening in the headless front and the serverless area and the whole front end movements with with react and pwa and view and next and next that's when we said hey there's a real opportunity here and we think there's a problem that's really unique that our customers face in the e-commerce space that we actually can bring something valuable to market and and that's when we launched this new product the xdn cool so we'll get there in a sec because i'm super interested in that part and this is also part of why i wanted to interview you guys but i want to jump into where mark comes in also because um i've been around startups especially in france for a while now and i've been looking into hey do i join one and what would that look like and then i said you know what because i come from enterprise kind of businesses pretty big ones agencies and stuff and a cto at an agency is quite different than its vp of engineering for example it's quite salesy with overview-like and managerial tasks and so between you and mark like you have a vp of engineering and a cto and in your startup how does that function what type of cto or vp of engineering are you guys maybe for mark what do you think yeah well we work very closely together so a lot of what we do overlaps it's just like sometimes it's just better to have two people available to do a thing like i i jump on sales calls all the time as does each one yeah both are involved in the product management um where we did for a little bit as i i'm a bit more just involved in engineering operations so i have the engineering team reporting to me uh i manage the stand up i do a lot of the backlog grooming and then i'm also much more hands-on in terms of the actual engineering work and just helping people you know get unblocked reviewing prs all that kind of stuff okay so it's a little bit more day-to-day and making sure the ci stack runs smoothly and all this kind of stuff yeah exactly so what is left so for you ishan is that more like figuring out new road map ideas or just you go away for a week you hack something together and suddenly you have a product how does that work yeah so it's you know i actually often reference um vogels the amazon cto and he has an essay that it's some well-known called the four types of cto and he breaks it into the the infrastructure manager cto the tech visionary the operations manager the external facing technologist and kind of like the big thinker um and so my role is more on the product management side and i do a little bit more of the customer facing side of it and prospect facing although we we often will pull mark in because look we're a really technical product and accepted go through webpack and really debug the hell out of that thing uh and so we'll often pull him in as well but what i would say is so i'm more on on kind of the the user and customer facing side and more on the product management side um and the way to think of it is between mark and myself with aj our ceo who's also very technical we kind of cover all of those four areas of the four types of cto together uh oh cool i for some reason it just came back to me at one point when i was still in amsterdam this is years ago i actually had beers with werner vogels from amazon oh really i didn't know him because he knew my friend who was like an agency she had been in agency for years and he had just started this idea of actually making those servers like aws he they were like six months in sorry that just came to mind and like i had no idea we didn't have cdns back then we didn't have companies like move web because he didn't have the art the the infrastructure right but maybe this is actually a good bridge into you guys maybe to explain a little bit what move web is like maybe both of you can try to say it in like one sentence uh yeah so i'll i'll i'll try it first but the xdn is a platform that enables sub second site speed for large database driven sites while making it simpler to manage your headless modern front end you must have set that one a couple of times yes that's why that's why i was ready to go and i'll give it a quick i'll just i don't know if mark wants to take a stab well maybe i'll just take it from a different angle yeah sure technical angle so it's it's serverless infrastructure for hosting modern web apps uh particularly single page applications built on react uh view angular etc okay so the point is that it's not jam stack right there are actually dynamically rendered pages that you might okay yes both so one of the problems that we see in the market is this we focus on e-commerce and you get e-commerce stores that sell five products and then you get e-commerce stores that are like auto parts dealers that sell a million products and so a purely static site works well for the five product company not so much for the auto parts company and there's a bunch in between and so there's only a few of us in this space that are allowing users to build apps that are partially static and partially dynamic and essentially static falling back to dynamics so it's it's a real world solution for sites that are just immense in size yeah so we've seen this issue quite often because i work with these kind of huge [Music] clients right mainly luxury industry but also different where we have like a hundred thousand products and so we are always like okay we're going to make this static because we know we don't have the security issues we can scale it easily we can deploy easily all of the stuff right but then we're like yeah okay but if you have a hundred thousand product pages and a whole bunch of other stuff how long does that deploy take and how wasteful is that for the resources of the hosting provider right and then we always put it next to yeah but we have a site core or we have a adobe aem stack how long does that deploy take and it maybe takes half an hour or something but i'm guessing it's like there's always things to put against each other but we haven't really found a way to make kind of this dynamic app and then have part of it static because of hcdn which is maybe xdn right so what does xdn stand for exactly experience delivery network oh fancy so so yeah but it's a deliberate player on cdn um and what you're talking about is is completely you know accurate and very astute and mark and i actually gave a talk at uh nexus conference last year about this this whole like build time friction and we went through some stats like the average you know physical store has even a couple thousand to tens of thousands of products in it and the physical walmart which is a us reference you know has over a hundred thousand products um and we showed a couple of ecommerce stores we have customers who have ten thousands of products we have customers who have over a million pages wow um and then if you look at content i we had some stats up there for like you know new york times and washington post they're pushing i think it was like 50 000 new articles per year right so imagine that to make a change if they rebuild all of that and if you're in a large e-commerce company they're people whose job it is to like move their merchandisers right they they're changing categories in the copy and the properties of products things go in and out of stock all the time and if you have to rebuild the whole experience like and if your build times are going from like 5 minutes to 10 minutes to 20 minutes it really slows down your velocity so this is where we felt we could bring a really differentiated solution that takes the benefits of jam stack and brings them to large dynamic sites and and the cdn's kind of the linchpin in that and to me jam stack is a spectrum um there's there's purely static and there's dynamic and there's a variety of jam stack like architectures but the key thing that i think they all they all have really is is two things one is front-end developer empowerment like the ability for a front-end developer to get as much done as they can without having meeting devops or back-end engineer and then the second thing is bringing data to the edge taking your your pages and making sure as much as possible can be served by the cdn and what classic gem stack does when it forces you to write a static page is it's forcing you to write that page in a way it could be served from the cdn but that doesn't mean that's the only way to get that that fast speed that you get with jamstack you can take a dynamic page and you can get it served from the cdn but you need more sophisticated cache management and so that's where we've added a lot of capabilities so you can fragment your cache you can do a b testing and your cache stays coherent um and and do things like micro target and micro clear your cache when say a product goes out of stock you only rebuild that page and then we've done other things like uh parallel static rendering which i'll i'll let mark talk about which are really interesting and bring the same kind of jam stack benefits to yeah man i wish i knew about you guys when we did the architecture for one of our like i cannot name the client they're one of those where we cannot talk about it but they're a huge e-commerce also in the luxury space and we just moved them all to a headless system because they used to be on a monolith and it was like 12-hour deploys that would go wrong all the time because it was a 15 year old system right so we we managed to build like an api layer in between on our end in that old system but then put a completely separate front end there that would just do ajax calls basically yeah and we we asked them okay so are you gonna do maybe azure or something no no we have surfers uh in germany in the basement like oh okay oh there are no gs i don't know like one of those right so we are not really like you guys we don't we didn't really know that much of that detail so we managed to have a bunch of servers up and running it all works but we made it so that it has if it it they only need to have 10 hits a second to the origin and it's fine because the rest is just hard coded in the cdn right so it's basically what you did in a really crude way because now for us what's really hard is to actually manage that stock manage the search results if they want to change them up a bit like those search results based on the urls are stored in the cache a certain way if you want to change them five times a day we don't have like that micro change that you just discussed oh man i wish we had move web you probably existed we just didn't know yet wow you said something really though uh a really great observation it's like you were able to put something similar together well we tooled it together ourselves basically yeah you told it but it's the maintenance and to have something that worked reliably and to do it day to day like you could set this up where you people will end up struggling when they try to diy it is it's on the maintenance side and that's some of the stuff we take off of people's shoulders so they can focus on you know really differentiated features and not just tooling and just keeping servers up it's a huge part of the value proposition yeah so mark how do you guys do this technically are you using like background functions like lambda function that can run for a while to maybe just statically generate something so so ishan hinted at that when he talked about a feature of move webs called parallel static rendering so we use lambda you know serverless functions to do all the server-side rendering on the platform and when you think about it statically generating a site is just doing a bunch of server-side rendering at build time instead of at runtime it's the same process and so um one of the problems with statically generated sites can be getting a deployment out on a large site could take a long time especially if the framework itself like like some frameworks are slower than others um and so with parallel static rendering we can actually push the site out so it's live immediately and then we use the um concurrency of all of our serverless functions we have available to actually hit the site and and server side render every page that you want server side rendered and so we can do like a thousand or more pages at a time and really rip through the site quite quickly but even if a user you know hits a page while we're iterating through all those the page will still be served and the first time it's served it'll be put in the cache so that all the other users get it instantly from the edge cache man this is so small we leveraged that kind of a you know highly scalable serverless infrastructure yeah it kind of flips jam stack on its head instead of exactly and deploy we deploy and then build yeah man because we've had things where we said okay we're going to have to go fully dynamic and we cannot do a cdn on a lot of those pages and we had a react app at that point and the ssr over react app is just hard look at walmart what they had to jump through hoops to get it to work right we tried also but actually that website it was so much faster than what they had before that they are happy but i wasn't happy because we couldn't do the on-the-fly ssr and then just cache it like you guys do so this is an an interesting one here like so we have the big guys like netlify and for sale starting to grow also how come you guys are not named in that same um like amount like you have netlify for cell to do this but where's move web in there is it because you're just slightly different or you're you're not competing on that on their plane or how does that how does that work with you guys yeah so we are newer so at least the product we've the company's been around for for a longer time helping specifically e-commerce folks sure and the other thing is we're more focused so we are we are very focused the the platform obviously is vertical agnostic you can use it on any side and as mark said earlier you can use it for a small site that's static or you can use it for a large dynamic site but where the differentiation of the product hits home is when you're this very large e-commerce company um or somebody who's got a very large database driven site and so we have a much more focused kind of go to market at least uh in the next 12 months or so that we're really targeting and that's probably the two reasons that we're not as well known but um you know we're we're working to get ourselves better known and you'll see us sometimes listed on you know uh if you go to like say the the viewer or knox deploy docs i i think we're also listed as one of the possible deployments yeah you're at next i know yeah yeah so uh we're we're definitely you know trying to make sure people know about us but our game is a little different i think in terms of the goal we're playing for is not to get as many developers as possible we'd certainly like the awareness um the people we're really focused on and our our our strategies really you know be really focused on a particular customer in a particular segment and those unique problems and that's what drives why our roadmap so unique compared to the other guys yeah because honestly i am your target audience yeah like that's why i'm so happy when we had a call a while back about all this stuff i'm like oh this is perfect because i am a tech director fro of front-end globally at an agency with huge clients where we want to use new technology but we are sometimes stuck and we have to try to sell it to clients and we help together and work on it but if you have a partner like move web it's so much easier to sell that new stuff in because they don't have to go back to the sales force right that just has one big chunk and it just works because you guys this has kind of a bridge to the next part because you're like part of the mac alliance right yes and so this is kind of funny because you it's to me when i saw because where i work at fall tech we also helped co-found the mac alliance and we talked about it internally and we actually said it's like there's all this new stuff out there and we want to work with it but it's so hard to sell in to a huge client because they're like what if this startup just falls over or they pivot then we just invested a million or five million and then we don't have them anymore so in our minds starting an alliance like this just make sure that we have more trust around is this part of why you guys are there or is it maybe also the visibility or what what's your mindset um in the alliance it's both i i think it's very much what you're talking about it's the fear of the new especially in larger companies right these are these are purchases quite frankly that um you know you're betting your career on oh clearly especially if your boss doesn't like what you chose it's kind of done yeah exactly if you're the you know you manage the e-commerce site for a very large substantial you know e-commerce company your career is on the line when you bring in when you bring in a technology stack or a platform and there can be a lot of fear around that uh you know i i knew an enterprise sales guy he used to say you're actually selling promotions um you know if it works people get promoted if it doesn't they they they end up getting accredited wow um and so there's a you know in an enterprise context those things really matter so what the mac alliance i think really does it helps promote this new thing reduce the fear and i think there's also a uh dilemma of choice that you have when you're going headless there's definitely a movement to headless it gives people when done right better freedom better velocity and development and better site speed but when done right and when the right components are chosen together so there's this like what are the components i put together and then what is that right architecture look like so i actually get these results that are promised because a great analogy is microservices we saw a bunch of people go through microservices which you can think of as you know it's kind of the precursors it's the m and mock alliance except if you didn't do it right you ended up with spaghetti that was oh man yes i've been there exactly and so for us the education of the market on the right way to do it is really important and and we think mock is going to play an important role the mock alliance and we want to help make that happen and and we think we're a really important component part of i think you are head back yeah we like to say where the neck for your headless exactly it glues stuff together kind of and i love that um so i'm going to piv it a little so mark you are um vc of engineering right so i can imagine in a startup where deadlines are probably a real thing there's a pretty big team there's complicated stuff you're doing how do you deal with people and how do you make them happy developers i think developers are happy when they feel kind of personal investment and ownership in the features they're building and in that regard we kind of have an unfair advantage in the space that we're in in that we're building tools for developers we're building a development platform and so as a developer almost any other type of job you would take you're just a developer when you work for a company like mova that's a platform uh company you are also a domain expert which brings its own challenges sometimes you actually understand the domain way better than your product manager would or your founder or whatever and so it's actually pretty easy and i've worked for two of these types of companies now and developer satisfaction is one of the last things you have to worry about at these types of companies because you're so empowered to use your own creativity experience to create products that really resonate with you as a developer so it's like cheating it's a very easy one oh wow i love that so that also means you're kind of not really there's not real hierarchy i can imagine right there's a hierarchy like you know some of us have more experience in some things than others sure you know and we work with a very sophisticated set of technology clearly um you know and use our kind of wisdom and experience um but in a way everybody even a junior developer is like at least a mid-level product manager in this regard yeah because they really understand the space in a way that you wouldn't if you were working for like a pharmaceutical company or a bank um you have very little understanding of that space but as a developer you've used developer tools before you you gain your knowledge of the space every day that you work yeah so you like where i work for example there's there's a more clear hierarchy that might also be com like company and country based i mean france where it's more like that but what i always try to do is like what you said right we're trying to make people feel responsible and take ownership of the stuff they do and then suddenly that hierarchy doesn't really matter anymore because it kind of boils bottom up yeah i guess if somebody who is a junior developer but loves something they just take it and then just run with it and suddenly that becomes the norm and so i actually where i work i had to make that happen but i'm assuming that for you guys this is kind of automatic right it's easier yeah it's easier because developers can very quickly find things they're passionate about whereas my first startup i worked for was a company in the in the biology space life sciences space and it was very difficult for developers to understand what the heck they were doing uh yeah understand the business and so motivation was a real problem there but here you know a developer even if you're junior there could be one very particular part of the space that you understand really well or that you've had pain with in the past and so you'd like to solve that pain on behalf of other developers so it's very easy to find motivation i like that so as you're someone in your job what what are the the harder bits what are the bottlenecks in that case in my specific job one of the things that is challenging is i've been in this type of a role for a while and in other companies i've actually felt like i knew the most about almost everything technically um i may have been more way more senior than other folks it's kind of beyond my skill set to be that at move whip given the wide swath of things that we have so i'm i go really deep on front-end technology but servers and cdns and networking infrastructure i wish i could be the expert at that at movement i'm good but i have other people that i have to defer to so knowing when to to defer and and when to take a leadership role on very specific technical points is a challenging thing and when you're responsible for a lot of things you feel like you should be the expert in everything and be able to solve every problem and at some point it just becomes impossible yeah and so you have to take that leap of faith and confidence in others yeah i must be honest i felt that too but added as an agency right it's a bit different but thanks for those answers it might make me want to move to startups it's a very interesting kind of approach it's just a different world right so guys i really want to thank both of you to take the time out of your busy schedules i really appreciate it and also i'm super excited to see what comes of move web and also what starts to happen now that you're in the mac alliance that more brands might use these services i most certainly will because i'm super interested so uh thanks again for your time and um yeah i hope to speak to you soon about some collaborations yeah thanks for having us yeah thank you for having us awesome guys thanks for your time cheers bye