CTO of OpenTable Talks AI & Tech Leadership With Cormac Twomey

Intro tell me a little bit about what what you're seeing that's happening and what some of your predictions will be for the next five years the first thing I would say is be really suspicious of anybody out there who's making predictions about where AI is going when tech companies make massive changes to their Tech Stacks do you ever wonder who makes that decision and why I sat down with the brilliant Cormac to me the former CTO of OpenTable and former director of eHarmony where we talked about his career Journey from being a software engineer to becoming the CTO of one of the most successful tech companies in the industry and the difficult decisions he had to make along the way both Technical and people decisions we also talked some AI just to scratch your itch let's dive in good morning everybody Welcome to the tech leaders Playbook I'm Meet Cormac extremely excited about today's guest Cormac I've known Cormac to me for about I would say 13 years Cormac something like that yep about 13 years 13 years Cormac was a client of mine and he was introduced by a friend of ours A Carlos Moreno who who told me at the time I'm going to introduce you to someone amazing I thought that was awesome and and uh Cormac and I hit it off um and since then we've we've remained as uh as buddies and I'm thrilled to have you thank you so much for being here Cormac yeah no I'm delighted to to be here so uh looking forward to the chat thanks for taking the leap of faith I know we we did zero preparation we just kind of showed up like friends would and to to have a good conversation so I'd love to start with just just your story your journey uh a cool cool thing about Cormac is he's been programming since he was eight and even though he was a CTO of a organization a publicly traded company an OpenTable which was part of a publicly traded company uh and managed I believe 400 Engineers yeah that's right if I remember correctly you still code you're still very technical and Hands-On well I I don't get me wrong I I like to code occasionally I have written approximately zero lines of production code at open table um so I I try to keep my uh coding activities now for kind of social purposes are you know on the side um I'm not a big believer in the CTO or head of engineering you know rolling up their sleeves and and taking you know production coding responsibilities probably best to leave it up to the so people are doing that full-time smart you've been programming since you were eight how does that maybe their Journey How Cormac got into programming starts there tell me tell me your story how did that happen how did you get to where you're at today you know it's funny my um well first of all I have my older brother to thank for having twisted my my mother's arm to get us a Commodore vic-20 um which I think was right around 1980 and um uh we you know we got that into the house my mom really wasn't sure you know it was just like he start you know articles are already showing up about about things that just suck kids further into screen time and uh she spoke she chatted to a friend who was in a computer science faculty in in Dublin uh where I'm from and um he said you know just buy a couple of games you know get you know give the kids a bit of fun through the Christmas holidays um but they'll get bored of the games pretty quickly and one of the nice things about the old simple computers that you could get at the time was that right out of the box the computer came with a program manual oh and so you know you turn on the computer there's a prompt right there um just loading to play a game is actually a is actually a command you have to enter you know you load from tape um and that would you know load up your game so pretty quickly that that played out and um you know we we played these couple of games simple games they were a lot of fun very addictive but it only takes you so far and I started like looking at a little bit more about at this basic programming manual and um you know what's funny I have chats with my son sometimes about this and uh my older son that I almost feel like I had an advantage growing up you know sort of like becoming aware of computers in 1980 because if I got my name printing on the screen like and then got it like printing you know 10 times and then moving left to right or something I felt like a programming genius you know and Meanwhile we're growing up in Silicon Valley and um you know there was a like three or four years ago there was a computer day where people were like showing off some computer stuff and as a guest speaker to introduce the uh um to introduce the this the session with the guest speaker was like somebody who had had a hand in inventing TCP I can't remember the name of the name of the guy but you know a very you know prominent name and and I I almost feel like it was an advantage that I had that there was nobody around me at the time who was just you know rocking the world of computer science and so if I get my name moving across the screen I felt like a genius and that gave me the encouragement to keep moving forward whereas I you know I see a lot of younger kids now who will do something amazing on the computer and then they'll turn around and say oh I'm no good at computers I don't know what I'm doing uh the bar just seems so impossibly high so um there's something to be said for keeping the bar low keep your enjoyment high and that certainly fueled me through my teenage years I just loved coding it was like solving it was like solving crossword puzzles on the Fly and you know getting some sort of nicer result than a crossword puzzle being done you know and then you go off to college eventually later and and study computer science that's right yeah I decided I Studying Computer Science don't know I think I've decided when I was 11 that this is this is really what I wanted to do when I learned that he could get paid for this stuff and um uh yeah I went off to I started computer science at the University in Dublin Dublin City University um and as it happens the the head of the Department of computer science there was my mom's friend from back in the day that had nothing to do with me getting in it was the Irish admission system is very cut and dry you like you have SATs the equivalent of SATs at the end of your school years the end of your senior year and whatever you get in those exams determines your future uh in terms of the like the scores you get the grades you get that tells you did you get into the course you want to take um anyway so I got into DCU uh did computer science there and just just loved it and then I'm guessing you started uh in Tech in some kind of uh Software Engineering software engineering role I'm guessing right yeah so when I was uh one of the one of the things about uh doing computer science in DCU is that um they're very focused towards internships at that University and so they the the year breaks early in your junior year and so I had a six-month internship and I was lucky enough to get get a job offer from the company in Santa Monica uh not a company that exists today it was called Quarterdeck but they were doing some cool stuff at the time and um you know I went over me and a buddy just two of us from from DCU went over and uh had an amazing summer living on Venice Beach and going to work every day it was it was it was such a great start into into the career getting paid to do something fun it felt to me like the you know the the rest of the department the engineering department headquarter deck they were just Geniuses and um just learning a mountain of information from the folks there it actually um I mean to change gears for a second it when I look at post covet now and everybody's working from home um working from home is great by the way in my in my experience uh a good team of Engineers is not going to work slowly when they're working from home if anything output goes up um it certainly is no worse but there is there is one piece that feels like it's missing that environment I was in when I was a junior engineer both when I was interning but also as a new graduate um I was lucky enough that this company gave me a full-time job offer job offer for when I graduated which I which I accepted gladly and um that environment I was in you know in in 1995 I was kind of you know one year on the job and I was I had the opportunity to to take the lead on a you know back-end piece of software which I was super excited about one day I I was I was literally over the water cooler like you know so classic water cooler conversation and another engineer who I you know I just sort of knew to see in the department was in there and he just asked me about this project and how's it going and I was full of enthusiasm and he asked me how you know what I was doing to handle a high degree of concurrency you know and I proudly declared that oh I had a separate thread for every incoming request this is the best way to go and and he he said to me like are you sure that's the best way to go and I said oh yeah yeah totally that's the way to do it and and he just looked at me and he said it's interesting you think that and then he left the kitchen and that comic just got under my skin you know and this is 1995 was before Google it was barely it was before Alta Vista even um and so you know I go into the the you know the the library the set of books that we had and you know in work and I'm looking through stuff and reading up it was in Windows NT I was looking up Windows NT and I was learning all kinds of things and I realized there were like there were smarter ways to go just because that engineer had said it interesting you think that and left the kitchen you know everybody I know who is a strong engineer now who has you know decade plus of experience they all want to ask them about their early days of their career there's always some interactions with other Engineers that that gave them an input that they didn't have you know and so and that's one thing so I worry about in the post covert days you know let's make sure we're maintaining that type of atmosphere I agree completely Cormac it that seems like an invaluable piece of knowledge that you got from a potential Mentor early on in your career that you would not have otherwise I highly doubt he he would have slacked you this obviously there was no slack at the time I highly doubt he would have taken the time to meet with you on a on an email or you know some kind of video meeting and that's I agree with you 100 it's missing it's missing uh especially with young folks I think if you're a 20-year veteran and you've been programming for 20 years and you have a great setup and you're disciplined and focused you probably can get a lot more done at home I agree especially in programming right in certain industries is not as as easy to be productive I think remotely but I think who's the people missing out the most are the young people and what I'm realizing it's the young people that are wanting demanding the remote work and it's I genuinely believe it's hurting them um I see more and more folks that want to get out of the house and at least work in an office two three days a week yeah just so you can get that camaraderie back that get out of the house it's it's healthy mentally um so I'm glad you said that because it's become almost um odd to ask people to work in an office and I think it's really hurting uh Young Folks young Engineers young you know sales people recruiters uh business people I think it's really hurting us yeah I mean I I think so too I think two to three days a week feels like a great sweet spot to me um you're getting you're getting that different Dynamic you're getting a little bit of sort of the the you know the unofficial information share environment in an office um you're still holding on to a great deal of flexibility in your in your work schedule if you're spending you know approximately half the time working from home um so that that's certainly my my you know my sweet spot I think pushing people into the office five days a week is um personally I think it's Overkill I think let's let's respect that everybody seems to be working happily from home they're working productively from home I think if you're if you have a team that's not working productively from home there's a bunch of other things that's going wrong with that environment it's not it's not that people are lazy I think when people feel challenged and they're set up for success in my experiences people respond to that and they lean into it agreed very very Keeping People Engaged good segue into uh where we're at next so Eddie Harmony uh I remember you had your team on site yeah uh that you know everybody I remember the Pasadena Office the Santa Monica office um how did you go about keeping people engaged how did you go about accomplishing massive projects uh while maintaining what other people wanted to accomplish because the individual has their own career goals the leader like yourself at the time I think you were a director of engineering right you had your own goals and the company goals so how did you align those those three things together their goals your goals and the company's goals look at myself first and foremost I'm a technologist I'm an Engineers engineer I My My by default I'm going to um admire great technical work I'm going to coach technically um so that sort of comes without without effort to me so having said that having you know knowing that that essentially is part of the day-to-day conversation regardless um I spent a lot of time emphasizing the focus that we exist to deliver business value to our customers and we are going to day-to-day evaluate our success based on meeting meeting customer goals and so you're you're trying to do is deliver you know you're trying to Delight your customer with a certain velocity you know how quickly can you deliver new uh initiatives and new functionality that's going to Delight your customer um now I think everybody can agree on that that's valuable trying to do that as quickly as possible but over what time Horizon are you measuring velocity if if you tell me that look I need this widget built in the next two weeks otherwise the company's going under well in that situation there's clearly a sort of there's an existential threat that we have to rise to that occasion and when there's an extreme time pressure we are going to sacrifice every engineering standard in order to meet that goal to say if you need something in record time we can deliver it it's not going to be pretty but we can deliver it um if you keep doing that from one two weeks to the next two weeks the following two weeks you're going to end up creating so much trash in your code base that you will be sprinting through tar pretty soon and you know you can be the biggest genius in the world greatest engineer ever but if you're sprinting through Tire you will tire out through tar sorry it will tire out very quickly so and and we'll all be going slowly and so while we keep the North Star as delivering customer value as quickly as possible I like to have a Time Horizon that's measured in the 12 to 18 month period something something around that which is to say we're trying to deliver uh functionality at velocity every month we're pushing out new functionality to our customers but I want to feel in the back of my mind that the stuff we're doing is going to result in 12 months from now 18 months from now we'll be going even faster than this I always want to tell me that in the back of my mind um it still allows for an occasional two-week period where we have to throw our hands in the air and say this we need to do this thing get it done right now but if we're if we're keeping that 12 month time Horizon in the back of our heads then as soon as you've delivered that feature you've risen to the occasion and and you've you've made something happen you need to say okay let's be honest with ourselves we create we broke a few rules here we've created a few bad patterns we you know went around some of our design principles our apis you know structure let's get that fixed um and get the right pieces in place so that this is going to be sustainable and this is this is why you're in the position you're in I always remember you you were a real you were a realist but Technical Debt you were a how do I say this you were really really focused on great code and great standards and great operating that's why when you used to hire Engineers you went about it very differently than the norm um Cormac can I guess that the the the statement you made about people just kind of throwing spaghetti code every two weeks just releasing is this how technical debt comes comes to be because it's a very common thing technical debt is greater when you are throwing a lot of code quickly as quickly as you can as frequently as you can then technical debt builds up very quickly but every single engineering team in the world has technical debt okay so you can you can have the greatest engineering team um you're still going to have technical debt because of what is the definition of technical debt is there anything that I can point to in my code base that if I change that thing it might speed us up a bit more the answer to that question is always yes there's always something that you could do to speed up now is it always worth investing the energy in in optimizing that at some point the answer the answer becomes no um or it becomes less frequently if you are a team that is staying on top of the stuff that's really holding you back then you end up you get to a point where you're spending a lower percentage of your time fixing technical debt nice Cormac you go into a company and uh and How do you deal with spaghetti code they told you all the right things but the first thing that they tell you is we got to get this out in the next two to four weeks and you've got your standards you've been very communicative about the fact that you don't throw out spaghetti code and it's all about long-term uh scalability and uh documentation and and clean code how do you deal with something like that because we have a lot of ctOS that go in the startups and they might not be around in 12 to 18 months they're not worrying about 12 to 18 months how do you deal with that I don't believe I said I don't put out spaghetti code if there is the first because my first answer that question we've got to get this thing out within this month my first question is why okay and if you have a convincing answer as to why then come hell or high water like if it's if it's humanly possible we'll get that done but I will I will get it we'll get it done and then we will keep in mind the rules we broke or the mess we created and we will get that cleaned up earlier in my career I've been really lucky working with some brilliant people uh over the years I worked for a startup uh that was called trivita back in the 90s um uh Brad Allen was the CTO who's a very close friend to this day and just did an amazing technologist um and we created we created a data prediction engine and in the early days after the three founders I think we were just in the very early days we were a total of five people including the three founders and I was one of the two Engineers um and we had we had the you know the the the guts of this data engine built and the what was important was this back-end system and back in 1990 this was 97 I think there wasn't a lot of you know if you wanted to hook up a remote API to be able to talk to this thing then you kind of had to build a plumbing yourself and so you know we invested in that I built some Plumbing got a got a command line interface hooked up to it so that we could control this data engine and um but we had a we had a meeting with the investors two days later and you know Brad could appreciate that that the this thing was all controllable from the command line and that's fine we could see the strength of the engine working there but if you're going to show that to entrepreneurs to to investors they need to see something a little bit more uh visual than I think bringing up a command line and seeing some numbers spit out on the screen so in 48 hours I created a UI for this application um now some ux designers would no doubt probably poke holes in some of the visualizations I had but in 48 hours I got a UI created that that had the right charts and widgets and you know predictions done visually in a way that you could then present at an investor meeting so that the the right next Milestone was hit for the company in order to close that next round of funding so that we could go beyond being a five-person company so that's an example of you might want to build it right uh but if you build it right and it takes two weeks instead of two days then you failed so sometimes there are real reasons why you have to get things done and um I am not going to be so um precious about living by engineering standards every single day to say that no I'm not going to give the business what it needs in a three to four week period or whatever that time period might be but you can bet after the fact that after creating a two-day UI we kind of went back and did a little bit of redo work after that right because if you don't you are going to create a mess and you will slow yourselves down chromic how do you it's an interesting follow-up question to that um we all and I know you do we all love Worklife balance giving our teams good work-life balance right we want them to to be with their families I know you've got a beautiful family I think when I met you your your wife was carrying your baby at the time which I'm guessing based on my math it was your your youngest who's probably 13 now yeah that's right right because that's what we met we met around probably 2 000 or 2000 what year was it 2009 tens probably 2009 you do well 2010 would have been when my my my second guy was born so maybe maybe it was 2010. there it is yeah so so you have a family I have a family a lot of folks that we manage have families or frankly they don't but we still want to give them that balance however like you said certain times the business needs more output the business needs more velocity how do you balance those things out because I know you're a really caring leader but you're also very very about getting your work done and getting your business right and getting helping your company achieve its goals how do you deal with that I think in a healthy environment it begins with right from the CEO downwards it begins with a very clear Vision about what the company's objectives are um too often you see frequently changing priorities that's just going to create trash if you can avoid that if you can structure things in a way that from one quarter to the next there's a clear sense of priorities then I can turn around to my teams set clear expectations and when things are going right people don't need to work weekends and and all of that however every now and again there's a surprise and something ends up being more complicated than it was or a complication arose that was unexpected or whatever the factor is um we now need to deliver this urgently and so I like to leave it to the teams to figure out you know the how but when I look at my organization if teams are working in this I gotta sacrifice my personal time my family time weekends things like this if that happens more than once a year then things are very broken and we try and lead in and do something about that because if you in my experience Engineers will respond if they're if if if you need people to work late work long hours they will respond however their lives will become miserable and that's going to result in even speaking completely selfishly as you know as as somebody looking out for business value your your product output will suffer not to mention your engineers will at some point say you know I'm gonna go work something somewhere else somewhere that treats me better uh so we we try to maintain a pretty reasonable balance and in practice maybe once a year there's a time when people are stepping up and working a little bit harder but uh if it's happening if it's happening every other month then something needs to be we need to lean in and fix that situation that's great Looking for meaty technical challenges thank you for that um Cormac I remember if I can quote you from 13 years ago you you always loved meaty technical challenges you always talked about I want to solve crunchy meaty problems but only if it solves a real world problems right so as a leader what do you look for when you're evaluating this uh let's say you're evaluating a startup you're going to join although you're you're well beyond those years but any role how do you how do you measure the crunchiness of the technical problems but to make sure that there's real world problems that need to be solved which is what startups are trying to constantly do yeah so I will say my personal Outlook has has changed a bit since then in that my role uh back at the harmony um I was very directly technically involved in the The Challenge and the solution and it was amazing working with that team um some of whom I've still been working with uh in very recent years uh which has been awesome um but we were doing we we were solving a problem that had a very obvious importance to the end user uh I I don't think there are many jobs that are out there where you can say thanks to the work I did there are more babies in the world um you know the work we did as eHarmony at the time resulted in with with absolutely zero reduction in the quality of the compatibility standards that eHarmony made itself famous for back in those days um we were able to increase the quality of matches so that communication rates per match went up at by 30 percent which was huge um in other words more people were talking to more people which of course was leading to more long-term relationships and um that's something that was really satisfying as time has gone on my my particular motivation has to be has been defined interesting business challenges um and applying technology to an interesting business challenge I'm not particularly you know you you could you could offer me some new gig leading technology at the biggest say e-commerce operation in the world that wouldn't necessarily interest me because it's a very solved problem um so I'm you know I'm interested in um you know teams leading into a problem that if they solve it properly is going to put a smile on the end user's face um at OpenTable working closely with restaurants was amazing um when we were doing our job right and and getting the you know building software correctly we were making it easier for restaurants to deliver Hospitality to their customers which is the reason they're in business okay they um I don't think many people have started a restaurant business saying this is going to be the easiest way to make a book um every restaurateur has a vision for Hospitality of some kind that they you know that they are trying to realize and um one thing I've learned while working at OpenTable and working really closely with a lot of restaurateurs is the technology landscape is a nightmare for restaurants um not the individual pieces of software per se I think I think we've done an amazing job at OpenTable um in you know working towards solving that problem but there's a lot of restaurants that have maybe 25 30 different software systems that power their restaurant business and restaurants in general don't hire technologists uh to run their to run their restaurant they were they wrong they they hire people who know restaurants they know the back of house they know the front of house they know how to deliver hospitality and there is this spider's web of Technologies they need to lean on in order to realize their their dreams they you know the an obvious question is well you know they don't really need to run 25 30 pieces of software well the answer is yeah they do because those systems do exist for a reason and they do increase the productivity of the restaurant they you know by having uh software systems to manage your your staff rotations to manage your your food supplies you know your your Revenue your front of house and so on um your point of sales of course um it all will take aspects of the business out of the day-to-day and will help you focus on the business of hospitality but keeping 25 systems you know software systems talking to each other successfully is not a trivial matter and that's something that I think there's a huge opportunity there in in the restaurant industry to deliver a more seamless platform for restaurants but it has been a joy though working close to restaurants and working towards solving problems like that for them um has been you know and when when we can be successful there it's extremely satisfying because it's a very real world indication that the software you're creating is making life easier for for other people's you know day-to-day Consumer side of OpenTable business especially because I mean forget about the the restaurant side how about the consumer side as a as a consumer I'm big on using OpenTable because I know I could book it in seconds and don't have to call the restaurant to ask a bunch of questions and things like that I wish more restaurants used the OpenTable platform because I found it extremely easy enjoyable to use so I've always been a open table fan yeah they are awesome I mean that's something certainly that the that originally so it was was the the built the company identity was the ability to do this online is it's a personal productivity boost if you if um if you can remember back to the days before online reservations it can be extremely tiresome if you're if you're trying to look for dinner on a Friday night 7 PM how many phone calls do you need to make uh before you have the right the right reservation in your hands um so it's it's a it's a it's a it's it's a big productivity boost for the individual as well all right let's talk a little bit about technology Stacks I know you're a technologist to you Tech Stacks probably language is just another another means to solving a problem uh eHarmony was a Java uh Java shop if I remember correctly and then OpenTable when you joined as I believe you joined as a VP and then became a CTO um was a.net shop I believe so I don't know how much Legacy you had to deal with but tell me about your favorite tech Stacks uh Tech Stacks that drive you crazy and what are the crunchiest problems when dealing with uh with Legacy code or Legacy platforms yeah yeah you're right um uh when I started uh at OpenTable it was all.net and um dot net by the way c-sharp the the language that's primarily behind you know most popular on on.net is um is a really good language um it's Java in in the years that has been evolving has sort of got to a point where it's it's syntactically you know pretty much as nice as C sharp but C sharp is always I think had the lead on Java in terms of it just being slightly more more elegant language really back in the I'm surprised to hear that yeah I mean I I think just when you when you put them up side by side um c-sharp code is a little little you know more condensed easier to read and um nicer to code unfortunately though back in the day C sharp was only a Windows platform and um and that's really I think what made it unpopular um and certainly back in 2012 when I started at OpenTable you know Engineers were self-selecting themselves away from c-sharp so it was really challenging to hire great Engineers into a c-sharp environment now credit to Microsoft c-sharp is not in the same place today C sharp is now a platform that's available on a Mac on Linux across the board you can safely run c-sharp in the data center without you know requiring that all your boxes are running Windows so um that's something that I think I think there's there's uh you know a growing support for c-sharp today but at OpenTable we started a journey of of creating um split back end between Java and c-sharp um Java is not the sexiest of languages I'm I'm pretty aware but it definitely gets the job done and as much as I like playing with challenging Technologies I don't like to have the the technology footprint that is the day-to-day at a company shouldn't be too esoteric um so I'm not a big fan I'm not a big fan of you know making big investments in the more you know Cutting Edge languages whether it's rust or back in the day was Ruby which you know Ruby now is a little bit on the downhill um and that's you know that's the the the the problem you create for yourself is when you're making too aggressive about on the next generation of platforms um Javas was was a big big language back in 2012 and it's still a big language today it is yeah it is and when you're trying to create a stack that can be maintainable um I am I'm I'm I believe in the value of making boring decisions so you want to you would definitely want to to limit your your your technical experiments to just a few well one of the challenges Golang that we faced uh partnering together right because uh Cormac was nice enough to bring us into recruit for uh OpenTable at the time was you wanted the best engineers in the world uh but there was a challenge with that the best engineers in the world did not want to do C sharp and at the time there was no.net core there was no open source Microsoft so it really created a well what are we going to do and I think that's where you went with the Java mindset and we were able to hire some incredible Engineers who were willing to deal with some of the Microsoft uh Legacy you know applications uh while building most things in Java most new things in Java Cormac the the one of the more um kind of sexy up-and-coming programming languages that's been around for a long time but really didn't have a mass popular polarity is golang how do you feel about golang have you looked into it what are your thoughts I think goes is a great platform um I think that a company like OpenTable has a limited need for go the important for at least for the product engineering teams um the important thing is create readable maintainable business logic something that people can jump into and uh you know alter and refactor Etc there's a greater value to us keeping it in the language platform like Java um founder is in go I think go though is amazing as a replacement for you know classic low-level languages C and C plus plus and so forth uh it is terrific and some organizations I think have a very you know there's a very defensible need for a go when you've got core systems where you're trying to create um you know very tight performance and you have uh you're interacting with you know systems at a slightly lower level um I think go can be terrific so you know sometimes in the like in an operations team might have a good use for go uh for example but um but it's certainly not something that we you know really introduced into into the kind of production stock this the the the business software stack Cormac if you started a company tomorrow Business Software Stack uh I know it's a tough question to answer because it depends on the the what you're trying to accomplish but if you had to choose any programming language or any Tech stack what would it be and why well that's an interesting question I am not I'm my head is not enough space right now I'm not planning on starting my own business but uh nonetheless um I think there are a number of of perfectly fine answers that question you know um I could say um back end in Java front end and JavaScript um C sharp is actually a fine Choice as well you can build back end in Python which is not the most efficient way to build build software but I'm more interested in Engineers who can move quickly than squeezing every last drop out of CPU juice uh from a server um I think it's a fine trade-off to make to say we're spending 30 more on our server costs on our runtime but our Engineers are happy and they're moving together well what's most important is that you don't create uh you know a very heterogeneous stack with you know one piece running in one language another piece running in another language and another piece running out of third and fourth language that's that's a mess um but I do believe I do believe in not getting too experimental with languages there so I'd be looking at uh C sharp Java or python probably for the back end and um and I think JavaScript for the front end I don't like JavaScript uh it's it's an it's a messy language broke every I think language design rule in its creation but it is universal on the front end it is it does it's the one language that runs in every web browser and um you know with node and react um and there's a number of other you know great front-end Frameworks that I think it's a it's a it's a pretty good matchup for nice for the web stock wow what's your least favorite programming language well I'm coding up some I'm I'm coding up some Assembly Language with my son at the moment and uh on an old 1980s uh CPU which is which is a lot of fun but also infuriating as well but um uh but okay calling that that's a bit of a non-answer um the only language that I've really tried to Grapple with that I just hated in my career was Pearl and I realized I I know there's there's a close friends of Mines that that would hate me saying that because Pearl had you know back in the 90s Pearl had some Ardent followers but I really don't like that in Pearl there's always several ways of doing the exact same thing and there's you know there's if you get 10 Pearl programmers and you line them up and say just just write a little piece of code that sorts the list of numbers or something you'll get 10 totally different looking pieces of code and um when when when there's so much room for so much variability it's very very hard to create a large Software System that has a lot of design Integrity um and I certainly don't want to work in an environment where you've got to be the the police force of uh maintaining certain syntax style and and whatnot um that's kind of what I like about Java in that it's kind of a boring language and it you know assuming you're just following some basic coding maxims it's going to be readable by the rest of your Java colleagues um so yeah Pearl was probably my least least favorite language would you Cormac Who inspires Cormac would you believe I spoke to a really good engineer a week ago and he currently programs in Pearl as the company's main back-end application language I would believe that wow I would feel sorry for that individual and for that company but I would believe that I do know some of myself I was shocked Carmack tough question who inspires you whether it's an actual individual you worked with or it's someone that you look up to in the industry that you're you're kind of following in in their footsteps or or watching what they do I get inspiration from I think a lot of different sources um I would get inspiration on you know a weekly basis from somebody I worked with um in in my in my management style I'm I delegate a lot but I spend a lot of time interfacing with engineers and um trying to encourage a flat communication culture where anybody can talk to anybody within the product development uh organization um and one of the ways I would so the way I would cheat towards staying closely connected uh to the architecture and the you know the active problems that are affecting the engineering organization while delegating a lot is I would pay a great deal of attention to uh to Incident Management how incidents were happening however they were being followed up on how the post-mortems were going afterwards what the learning lessons were from those um and that would also create um you know openings for me educationally reach out to engineers and and and talk it through and the number of times I would get a little moment's inspiration from those conversations I think you know is just super super frequent um I over the years I've used Twitter quite a lot as well to be able to just follow individuals um in in within the tech space um they they being able to follow certain individuals it's like a shortcut just keeping on top of the technology ecosystem and evolving uh with that um it's been interesting over recent years Twitter I tried to just use Twitter like I say as this technical tool it it got more and more political 2015 2016 through recent years but I still find it super useful to to keep on top of technologists and and try and keep up with what people are discovering yeah I'm not going to call out a single individual I don't think that's I don't know out in the public space fair enough fair enough um I think the world will explode if I don't ask CTO about about uh current trends specifically in AI so let's let's get into that how do you how do you kind of keep up with current trends and the current Trend seems to be AI ai's clearly changing the world you and I knew that 10 years ago all right it's just it's just taking off 10x we didn't know it exactly in this way I I think it's fair to say of course not at the time I think you all were a very data driven uh culture and so obviously that bleeds into AI but tell me a little what about what what you're seeing that's happening and what some of your predictions will be for the next five years the first thing I would say is be really suspicious of anybody out there who's making predictions about where AI is going um last year was it in November the chat CPT was released to the public it was sometime around that enemy in the fall of last year and um open AI the company behind chat GPT they created it internally they didn't know what to do with it they didn't they didn't know there was any business there and they kind of shrug their shoulders and they were like well let's just release it to the public and you know whatever and we'll move on so you know if anybody was able to predict where AI was going I would point my finger at those guys right they were at the Forefront of of what we're all reacting to now and they couldn't quite see where where this was all going so you know if they can't I think a lot of the pundits that are out there they're going to the the you know we're all figuring this out as we go um that said um I think there is only one option for us in the you know in the near future which is we should be leaning into this technology just looking at how this is affecting software engineering my my immediate domain it is a the tools that exist today are already clear productivity boosters companies that are properly using the chat GPT and the tools based on those technologies that are already available to Engineers like copilot and other tools um that is already easily a 20 productivity boost per engineer and you know look at tools I co-pilot they cost 20 bucks a month they're so cheap um that we need to be leaning into that if you're not leaning into that then you are going to get left behind um I do think it is critical for the philosophers of the world and politicians and others to think about where this stuff is going um you know the existential threats and the Terminator types of conversations where people talk about a future run by AI is um those are important conversations to have but if we react to what's available today by saying you know by introducing restrictions on how we use today's generation of Technology uh for productivity boosts we will we will fall behind relative to other countries that do not have do not make those same decisions um this is a huge huge productivity boost um you know I was listening to an interview with Jeffrey Hinton recently um who was kind of like you know some people say the Godfather of of generative AI but you know he's certainly been working with generator of AI probably longer than anyone and he's you know he recently left Google um and he's been raising concerns about where the stuff is going and um he's describing the current generation of tools as an idiot savant right um and I think that's a pretty good label um the generative AI technologies that exist today are really really bad like really impressively bad at giving you truth right so don't trust the the generative AI systems to generate Truth for you um you know some lawyer some lazy lawyer out there found out the hardware I think and generate his legal Arguments for him and and yeah it didn't it didn't work out too well um because you know it created all kinds of it cited all kinds of previous cases that didn't exist and they were just it was just riddled with lies so if you're if you're relying uh if you think Chuck GPT is going to allow you to sit back and turn off your brain you're very mistaken in this this generation of tools you're very mistaken and and that's going to be the case for a while chat GPT 4.5 or whatever the next version will be is not going to suddenly fix that there's there's an architectural aspect uh to this and figuring out what truth is is a very very hard problem um so that's not something that's going to change I'm not going to say five years I don't I I wouldn't dare make a prediction that is as far out as five years from now in this rapidly moving space but um certainly over the next I'd say you know 18 months this is this is not um what you want to use tap TPT for however there are a lot of tasks in I think pretty much every domain but certainly in my domain in software engineering there are a lot of tasks that are more humdrum than others if I'm creating if I'm creating a piece of software um if I'm following good programming Unit tests practices and I'm uh creating you know compact individual pieces of code like I've got I've got a software interface and I've got a little class that's powering that interface I can show that to tools like touch EBT and say hey create the unit tests for me right sure um in modern I mean hopefully everybody is creating unit tests for their their code these days um chat GPT is not going to be a replacement for your brain when you want to create some clever tests you probably have to roll up your sleeves a bit still but 80 of the tests that you're creating when you're creating unit tests are very straightforward and that's just an example of Chachi petite can blast out a lot of tests for you tools are coming along I've spoken to a number of people who are working on these tools that are going to tell you very soon where your security vulnerabilities are in your code right you know there's there there's there's the owoss top 10 which is the the top 10 most common coding patterns you know essentially bugs in your code that allow hackers access to your stack right you know Avenues to it uh to enter your platform and um if if I could somehow have a tool tell me look this is where all your OS top 10 vulnerabilities are in your code that would be incredible you know to be able to you know and this is even before you have a robot that can say go fix them for me but if I could just if a team could just point themselves and say oh look there's three OS top 10 vulnerabilities we didn't know we had let's go fix them now that would be incredible and um that's a productivity boost it's also a um happiness boost doing this kind of stuff but you you can imagine the productivity uh loss you got when when you don't fix those problems and you get a nasty hack a year later or nine months later or whatever where some attackers get into your platform your business suffers um by the way if the attack is bad enough your business gets destroyed overnight there's I would rather have my data center blow up than have a really bad hack of my platform because you've lost all trust with your users when that happens and it's very hard to recover from that let's fast forward five years right so um companies are dealing with a lot of challenges uh as as many great disciplined Focus that people that are out there that work really hard remotely um there are as many stories of folks quite quitting not working people haven't turned on their computers in three four weeks and so on and so forth that combined with people don't want to work in an office salaries are highest ever even in in the layoffs and companies are reacting to that by having massive layoffs it's a correction right they over hired overpaid six out of ten companies said said they will have a layoff or have had a layoff in 2023 combined 2023 and 2022 we've had you know quarter million of layoffs just in the US from just big tech companies if it might be more than that by now so How to balance AI and tech leadership companies want efficiency uh financially and from what they produce out there and on the flip side most of the folks want to work less and live live more right how are we going to balance as Leaders how are we going to balance the needs of people and working and to employ people and there's young and upcoming computer scientists that are graduating and may not be needed I'm guessing manual QA Engineers are no longer needed right automation Engineers soon could be not as needed so how do you deal with how do you balance both sides you if you don't go with the flow and go and and utilize AI you're going to be a dinosaur and die off if you're full on with it we already know what's going to happen how do you deal with those two things and I don't know I don't know what's going to happen I do think that companies are going to become much more productive with their teams which is great but does that mean there's going to be lots of layoffs will some companies lay off Engineers yes but how will that balance out will that mean there's going to be a lower demand for engineers as we go forward my suspicion is that the answer to that is no there is going to be more opportunity this technology I think is going to create more opportunities for more for engineering in more places software engineering in more places um by lowering the barrier to entry to software as you just said to be this software engineering is more expensive than it ever has been the market is telling us that there is demand for this yes if you're one of the uh you know Engineers who is out of a job right now it sucks right now I don't mean to make light of that um I don't think it sucks as bad as it did in 2008 mind you this is this is something that is going to correct when you look at the the recruit recruiting numbers over the last three years the amount of Engineers that were recruited into the big tech companies so far when I look at layoff numbers it feels like we are correcting rather than going into sort of recessionary layoffs um yes it's awful if you're if you're one of the people affected um but I I do think that there's still uh active hiring it's taking longer if you're looking for a job today it's going to take you longer Is there still active hiring than it would have taken a couple of years ago sure um but I do think the market is telling us there's still tons of demand for engineers and we're busy there's there's plenty of hiring not as much as last year but there's still plenty of hiring going on and I think uh one thing that is going to be true that as productivity of individual Engineers goes up I don't think that's going to mean that Engineers are suddenly going to be making half the amount of money they were making before because my experience so far with the the generative tools is that you need to have a pretty good engineering brain to Pilot that those tools correctly you these tools will not make a sketchy engineer into a great engineer what these tools will they'll make a solid engineer into a much more productive engineer but you need to be a solid engineer you need to be reviewing everything that's created you need to be um you know interacting with the Tooling in such a way steering it correctly in order to get to that right output so it's an accelerator it's not a replacement for your knowledge though um I believe that's going to remain to be the case for a while again I'm not going to say five years from now it definitely will be this or that but certain I think three years three years is a horizon I'm comfortable with and I think in the next three years it's not going to it's not it's not going to be that replacement I I'd like to be standing you know at the cliff three years from now and looking out the next three years looks like um but but what exists today I think is going to be an accelerator and it's going to enable a number of organizations that currently look at the complexity and the cost of having a software operation and they're saying No thank you I think this tooling is going to create more flexibility for organizations to say maybe you know what I think if we can do it at half the price we thought we could do it at let's let's lean in and let's do it so I'm excited I think I think there's going to be a little bit of a an explosion of um opportunities for engineers actually Cormac let's um let's create a top five because people love numbers um what advice what are five things you would advise up and coming uh Engineers when they want to be ctOS or current ctOS that could learn from someone like you what are um what are the top fives you would recommend five things you would recommend to them well I don't know let me start with number one do something you love to do okay okay um now you were you already mentioned I started coding when I was eight by the time I was you know 11 12 years old I loved coding so that was figured out for me um but if you're in if you're in you know a computer science program and you're you're going to be the next generation of software engineer it's a huge domain with you know lots of directions you can go um make sure you are always challenging yourself there's a little bit of it when you're a junior engineer there's a bit of a conflict in terms of if you want to impress your employer you want to deliver on the business objective quickly and move on to the next challenge and then you know that's gonna be business Challenge number two business Challenge number three business Challenge number four and to do great jobs and impress your employer you need to move in quickly quickly quickly through all those challenges if you focus only on that it's going to be hard for you to turn into a great engineer long term you need to be finding time to deepen your knowledge to understand okay why are things working that way um why don't they work this other way what's going on you know in these Technologies I'm using let me learn some of the neighboring technologies that you know I'm calling these apis I know they work I don't know how they work there's my magic line um Steve yagi is a famous guy who used to work at Google and other places and uh he had a great um uh blog where he'd um you know have these posts but he talked about um I think it was called Practical Magic the one article that he he wrote and he basically said every engineer in the world from the laziest you know skirt scooting along just on top of apis to somebody who sort of knows everything every engineer in the world has magic lines and for some people that is you know I'm coding some web experiences I'm calling some apis the magic line is those apis I know that everything below that is just magic I don't know how it works it just does and now I'm like I'm pulling it together and often as a junior engineer you're kind of operating at that level um and then there are some other you know super deep Engineers who have learned all the layers through the operating system they've learned you know how you know garbage collection works and they've learned how you know UI Frameworks work how networking works all these things and um you know Steve yeah he'd like would talk about how at some point down at the microprocessor level how transistors work we're magic to him you know so like he goes he goes pretty low but the point is everybody has a magic line and you should always challenge yourself to push your magic line lower every year are you trying to deepen your sense of where that magic is by the way there's nothing wrong with starting out where your magic line is just that API you're calling I mean we all we all start there but find time and find the interest to figure out what's going on underneath the covers muck around on the weekend or you know whenever to be able to play around and learn something about this domain if you if you find time for yourself to understand those tools as a hobby more than how to get this current work task I you know I have in front of me to get that done um that's going to benefit you you are going to be the sort of engineer who will ultimately become a principal engineer and an architect um because when you've been doing that for 10 years a lot of the pieces start gelling together and you can you can spot patterns broader patterns you can um you know figure out what's going on in a much broader architecture stack when you have that deeper knowledge but if you're like I say if you're focusing on delighting your employer in your first job you know in exclusivity um then you might you might sell yourself short Cormac you know what one thing that I noticed uh one you talked about why when I asked you early on hey we need this project done in a month and the first question you had is why and then when we turned it around we talked about engineer forget CTO or VP a director um you talked about the why right because it sounds like if you love what you do then you don't you don't look at this as a nine to five just so you can check boxes so you can eat I feel like if that's what you're doing you're in the wrong business and you're missing out on so much debt you know they say that only 23 percent of employees are engaged or something like that at work and as bad as that is for the employer I think it's even worse for the individual because we spend one-third or even upwards of a hat if you don't count sleep at work so if you don't like what you do I personally recommend getting out of it and the reason I brought that up is because what you started with is do something you love because then when you geek out on the weekend to go learn and deepen your knowledge it's not it's you don't feel like you're not living right work life balance I'm working so I'm not living you're doing something you love and you're getting deeper and deeper in your in your world and when you do that you just gain so much right uh people always hey you work so hard and I'm like yeah because I love what I do if it's written there are so many other ways to make money I'm an entrepreneur you know that Cormac there's I do real estate I'm an advisor to companies and all these but I love what I do I love working with brilliant you know ctOS like yourself and Brilliant engineers and my awesome team here and so for me it's never felt like work and I work my ass off but I enjoy it so it doesn't feel like work I read a book a week because I love it not because someone assigns it to me so I love I love this this piece of advice this is great I I don't know very many people who I look at who have been very successful in life who don't love what they do because it's just it is so hard to devote the right kind of mental energy to a problem that you don't love um you know and I've I've been lucky software for me has always been fascinating and so you know I'm I'm a night owl so um kind of 10 P.M to midnight sometimes to 1am that's my time to sort of like to geek out um I still like doing lots of things that are out in the fresh air on the weekends yeah um I've just been being sailing actually is my my latest Hobby and uh that's been that's been that's been fun um so yeah I mean you know we all have to find our personal battles that make sense to us but it's I think it's really important uh that you are striving to deepen your knowledge it's it's a very interesting time with generative AI tools but for 40 years software Engineers have been making different business domains more and more productive and I don't remember being part of a team my career is not 40 years long by the way but um but um it's not much short of that but anyway I don't remember being in a conversation where we'd be saying I don't know if I should be doing this because it's going to make some other work domain too productive no making a domain more productive is amazing right and when we look at the unemployment numbers that we have today across the industry across you know the Western World certainly is its unemployment is low and it just shows you that productivity gains benefits more often than not it benefits what creative things we can get get up to as you know an ecosystem and that's why I think when it comes to generative AI we need to look at on this with optimism I have never worked for a company that had too many Engineers um I'm sure there's there are companies out there that exist but um invariably the problem is trying to figure out how to get more done how we've got as long ambitious business roadmap if only we had you know an engineering order that was twice the size of our current one but you know budgets are what they are this is an opportunity to to to fuel some of that ambition um so I'm I'm very optimistic about that great news is if we all get replaced by AI at least we could all become plumbers or or roofers I don't think anyone's doing that I think I think uh AI plumbers that's that's going to be that's gonna be a long time coming yeah yeah I'm gonna look into it I have a friend that owns a very successful uh plumbing company um Cormac what you've added tremendous value not to just me but the world after this hopefully what is one thing we can like what are you up to these days how can we uh add a little value to cormax World well um I don't know I'm actually enjoying life uh quite a lot um I recently left open table um which was uh you know it's a tough decision because OpenTable is an amazing place and the team is is incredible there but anyway after 10 years I've decided to move on and um I've been catching up with a lot of family and personal projects that I've been wanting to get to so uh taking a breather um I'm starting to talk to to companies now so looking for my my next opportunity but I'm you know I'm just interested in in uh meeting a business that is um solving an interesting problem for the customer that's it there you go there you go would you take on advise advisor roles board roles things like that I would um I I I an opportunity to be able to stay in conversation in a number of different domains would be very appealing so I I'd be very interested in participating in that capacity there you go folks you heard it you heard it live from Cormac himself Cormac you've been incredible I appreciate you greatly this has been so much fun you and I needed to catch up anyway we just happened to do it on online and hopefully I'll see you at the next uh CTO poker night yeah that'll be cool thanks very much if he does have really enjoyed the conversation thank you brother and that brings us to the end of another great episode of the tech leaders Playbook I want to thank you for joining us and hope you took away some valuable insights to apply in your professional Journey Don't forget to subscribe on your preferred podcast platform so you don't miss out on the next great conversation I promise it'll be good if you enjoyed today's episode we appreciate if you could leave us a review your feedback not only helps us improve but also help others discover the podcasts better leaders mean better working environments better working environments leads to happier people remember a rising being tied lifts all boats I'm Avida centavlian and this has been the tech leaders Playbook keep leading keep learning keep giving and I'll see you on the next one until then stay inspired my friends