Latest / CIO Unscripted / Felix Quintana on Becoming a People Leader and Managing AI Agents
Transcript
- Felix: for a long time IT ha was always considered they they're the introverts, they like to be down in the basement, they don't want to talk to people, they just want to be in front of their computer. but I I think this is where I attribute my business degree Niel: Okay. Felix: and some of the skills I learned in in in that process of you gotta be able to work with people. You gotta be willing to develop people and clear roadblocks for them and go have the necessary battles where where needed on behalf of your people, coaching, sometimes Niel: Yeah. Felix: disciplining, all of those things are are real really critical. And just because you've been a great individual contributor, it doesn't necessarily make you a good people leader Niel: Welcome to CIO Inscripted today with Felix Quintana. Felix and I have known each other for a long time. Felix has had a great career in technology and technology leadership. And one of the things we're going to talk on today is his he was doing AI back in the early days when AI existed of machine learning, robotic process automation, some natural language processing when it was kind of hard to do. So anyway, we're going to It's maybe it's hard to do it today as well. But anyway, we'll talk to Felix about that. again, I'd just like to thank VLCM for sponsoring this and making this possible. So with that, Felix, kind of give our audience an overview of your career, how you got started, how you worked your way into or how IT leadership was forced upon you. Let's put it. Felix: Yeah, thank you, Niel appreciate you having me on. it's always a pleasure to get a chance to chat with you. you know, I I didn't always want to be in IT. it similar to you, it happened to me. Early on when I s when I started my college career, I I actually wanted to be in international business. That that was the goal. at the time University of Utah didn't offer anything related to international business. So I I actually per started pursuing a marketing degree. Niel: Wait, wait. Now usually there's an incredible tension between the technology department and marketing. So I just wanna okay, so Felix: Right? Niel: so I'm glad you chose the right path after that, Felix: That's right. So my junior year of college, I we actually had a family friend reach out and said, Hey, I have this small internet service provider that I that I've been running for a couple of years and I could really use some help. would you be willing to come help me out? Like, well, what does that entail? And this was I I'm gonna age myself, but this was back in the days of dial up internet. It's like, well I I need someone to provide technical support, but we also do hosting and we also do some web design. So you would get to do all of that all of those things. I used the internet, you know, early days of internet, but I I I'm like, I don't I don't know that I'm the right person for you. Like I'll teach you. Don't don't don't worry. So so I went to work for for this small outfit in Riverton, Utah. And yeah, it was it was quite a learning process. However, every day brought new opportunities to learn new things. I was getting my hands in Linux and web design and dial up networking Windows 98 back in the day. You you you remember those days. Niel: I do. I remember the the squeal Felix: And and it wasn't it it wasn't like Niel: of the modem. Felix: Yeah, exactly. And it wasn't like, hey, let's get on a screen sharing session so I can walk you. Like you were literally walking people through setting up their computer over the phone. but I just fell in love with technology. That that's that was my entry path. and I went back to school and I said, I I need to go pursue something in in technology. At the time, the engineering school is the only one that was offering technology degrees, computer science, that type of thing. So I went and chatted with them and they said, You know what? we offer computer science minor, however, we give preference to those in the engineering school, and we only allow about five per semester. So you're chances are pretty pretty slim. So I I just went home devastated, but I I kept the course and the following semester the School of Business started offering an information systems degree. Niel: Okay. Felix: So that that was that was my calling. Niel: So you were like an inaugural MIS BIS student. That's interesting. Felix: I I was, yeah. I I was among the first group of students. so I still finished marketing and then I did information systems. Niel: Yeah. Yeah. Felix: People often ask me if my marketing degree ever comes in handy and I tell I use it all the time. I have to sell ideas, I have to sell the CFO on on approving projects. So I'm I'm I'm still glad I went and did marketing, but information systems is where where I pursued Niel: we we joke about the tension between technology and marketing, but you bring up a really good point. Every time I've done one of my transformations, I've developed a marketing IT plan. Felix: Yeah, you have to. Niel: I have to market our department. Yeah, both internally Felix: Yeah. Yeah. Niel: and externally. So it's interesting for recruiting purposes. Sorry for the interruption. Go ahead. Felix: No, no, you're fine. so I I actually stayed through through my college years at at this small outfit. I was there for about four years and then this online university that was in its early years called WGU came calling. They they had a need for for a network administrator. So so I went to work for for WGU. I was there for about seven years, you and I Niel: Okay, how big was WGU? How many students when that you first went there? Felix: When I first started we were twelve, thirteen thousand, Niel: Twelve thousand. Felix: some something like that. Our IT department was twenty five, thirty people at most. That included in software engineers, infrastructure, leadership, that type of thing. yeah, service desk. Niel: Service desk too, I'm sure. Yeah. Felix: We had two people in our service desk. Our IT Niel: Yeah. See. Felix: operations, which I was a part of, there were four of us. Niel: Yeah. Felix: manager and and and three three of us. So yeah, it it was it was fun times. At the time we were co-located at the Quest Cyber Center in Sydney Niel: Okay. Felix: Vale, California, occupying two racks of equipment to support All all twelve thousand students. One of my first projects was to build out a colo facility here in Utah. Sun Microsystems at the time had had made a pretty significant donation to the university. and we upgraded to about five hundred square feet of space. So we went from two racks to nine racks and we were in heaven, like net knew everything. It was actually really, really good experience for me to get. Like my first three months there, I was building out the entire infrastructure infrastructure with with my colleagues of course. So a lot of hands-on experience on in the infrastructure space and data center power and storage and networking and server management, all of those things. you came in at the tail end. Well I I should answer your question about how how I got into into leadership. I I was an individual contributor for about three years WGU and there were a multitude of leadership changes. You're you're familiar with some of those. at some point my my direct manager, the he was director of infrastructure, something to that effect, picked up and and left to to go work for Omniture at at the time. Niel: Okay. Felix: and that created a void. so the CTO at the time came to me and said, Hey, I see some qualities in in you, right? You may not have had the title, but I've seen some leadership skills and I think I think you'd be good in in leading this team. But I'm not ready to make it official. Let's let's put you in a center and basis and I still need you to do your other job. Niel: Of course, yes, yeah. Felix: that experiment lasted six months. and and it was interesting. I I had the pleasure of interviewing a lot of potential leaders that would come in and fill that that void after about six months. two or three members of the team went to the CTO and said, Why why are we still looking? Right? He he's obviously doing a decent job that you you you haven't fired him yet. So why let let's let's make it official. Niel: So did at that point did you get to interview yourself for the job then since you were interviewing your your potential candidates? Good, good. Felix: Yeah. I I I I pass with flying collars, Neil. Yeah. So so we we had a conversation said, okay, let let's make it official. Let's let's look at backfilling someone for for that network engineer role and and make it official. You'll you'll lead the IT operations team. interestingly enough, I I told him I still needed a few days to think about it. Six months had not been enough. like give give me the weekend to think about it. And honestly for me, I I really enjoyed being hands on keyboard, you know, Niel: Yeah. Felix: working with the technology and I really had to sit down like is this really what I want? But ultimately, I hope this doesn't come across as cheesy or or or or anything, but I had had a series of of not great managers, I'm I'm gonna say it, up to that point and and I felt this was an opportunity for me to maybe make break the mold and I wanted to be the leader I always wish I had. And and I had some Niel: Yeah. Yeah. Yeah. Felix: had some good leaders. I you know, not everyone was was was not great, but I saw it as an opportunity to maybe be a different type of leader and and I I embraced it and took that as a challenge. so I did that for about three years in WGU. You came in at the tail end of that, right? You and I had an opportunity to work for I don't know, six, eight months, something like Niel: Yeah. Yeah. Felix: that, towards the end of my career. but yeah, but yeah. Niel: So so let me ask you a quick question about that. You made the transition from technical leader to people leader. Felix: Mm-hmm. Yeah. Niel: And along with that was a goal that you're weren't going to be like the bad leaders you'd had, or managers, Felix: Yeah. Niel: and be like the good ones you either hadn't had or knew what you wanted. How difficult do you think it is? I'm just this is kind of an open question because I I ponder this all the time. How difficult is it for technical leaders? To pick up the people leadership skills they have? And does it it does it take a time and effort? I'm just asking, because in my case, and I'm not sure I'm a good people leader, but I had to work at it. You know what I mean? Because left to myself, Felix: You are. Yeah. Yeah. Niel: I'd never talk to another human being if I never had to. And now I had Felix: Well, Niel: to. So I'm just, I'm curious in your case, what was that transition like? Felix: that's a great question. And you're absolutely right. For for a long time IT ha was always considered they they're the introverts, they like to be down in the basement, they don't want to talk to people, they just want to be in front of their computer. but I I think this is where I attribute my business degree Niel: Okay. Felix: and and and some of the skills I learned in in in that process of you gotta be able to work with people. I'm you know, by nature I'm an introvert. so I like you, I I you know I I'm very selective of the human interactions. But I have but it's interesting. I I I've done those personality tests and and I actually have multiple personalities, Niel If Niel: Yeah. Felix: I'm at work, I'm forced to be an extrovert. I have to. Niel: Yeah. Felix: but in my personal life I I tend to be more reserved. it is hard. That transition from from managing technology to managing people is hard. And I and I think part of it is you have to sometimes people that are great at being an individual contributor assumptions are made that they will naturally make good people leaders and and and you have to want to be a people leader, right? Those that do it for the wrong reasons, either pursuing compensation or it's the only path of progression for them. Unfortunately that's not always the the the best reason to to go into people leadership. You gotta be willing to develop people and clear roadblocks for them and go have the necessary battles where where needed on behalf of your people, coaching, sometimes Niel: Yeah. Felix: disciplining, all of those things are are real really critical. And just because you've been a great individual contributor, it doesn't necessarily make you a good people leader So it it it is a mind shift for sure. Niel: Yeah, in my case, I had to give, like I said, give my choice, I wouldn't interact. And I recognized I had to proactively build relationships. So I had to, I had to go reach out to people and say, I I want to get to know you. I want to understand what you're trying to do. And it was I and I still at certain to a certain extent I have to remind myself, I don't want to do this, but I need to do this. Felix: Yeah. Niel: Okay, sorry about the interruption, but I think that's a critical one. Because I think a lot Felix: Yeah, sure. No, no, it's it's great. Niel: of technical leaders they that that they're com comfortable in the technology, but and aren't sure how to interact with the business, and they tend to refer back to the technology and technological details. Felix: Yeah. Niel: And therefore don't make a connection at the either the personal level or even the business outcome level, anyway. Felix: Exactly. in fact I I'm glad you brought that up. So af after I left WGU I went to work for for a shipping and logistics company called Data Two Logistics. And I still remember doing a a tour of of the data center 'cause it was hosted on prem Niel: Okay. Felix: with with the technical lead at the time and and we're we're doing this this walkthrough and I start picking up on things like, wow, like What what's that workstation sitting on the floor over there? don't worry about it. That's our NT server. This this was fifteen years ago. Like NT had Niel: Wow. Yeah. Felix: been deprecated for for a long, long time leading up. Like, okay, help help me understand. You know, I see a lot of technical debt here. What what's going on? like yeah, you know, there there there's just hasn't been a a strong interest in in modernizing. Like, well, that's gonna change. And I still remember her chuckling a little bit. Like, yeah, good luck. others have tried. So I took that as a challenge, right? And I I often like to talk about being able to speak the language of the business. Niel: Mm-hmm. Felix: within the first sixty days, I was able to put together a very strong plan as to here's all the shortfalls from a technology perspective. But I didn't go in talking about storage and networking and how you know. one gigabyte and we need it to go to ten gigabyte or you know i they don't they don't they Niel: Yeah, who knows what? Yep. No. Felix: don't understand that you you have to explain it so at the time we we were courting a a very important client that was actually threatening to to leave and I I remember going and having a conversation with the CEO at the time, like, what what what's going on? Do you understand why they're leaving? It's like, yeah, we we have this reporting platform that's supposed to give their metrics and it's supposed to be their first thing in the morning for them. our processing is taking so long that they're not getting it until mid-morning and and they're pretty frustrated with like, well What if I if I were to solve that problem with you? And and at the time, all flash storage was was coming into the scene. so I went to a technology partner, they brought in a an all flash array. We migrated their databases to it. and I still remember getting a call from this DBA that night, like, hey, I think something broke. I'm like, what do you mean something broke? Like, This ran too fast. it it's not supposed to be that fast. Like, well we have time, run it again and call me back. So he did and he like, Yep, it I've I've verified everything's working. This client is gonna be so happy. so sure enough the client was ecstatic and you know, for for weeks straight we were able to deliver their their their reports on time. CEO calls back and says, What's it gonna take to get every customer moved over? Well, Niel: Yeah. Felix: I gotta lose the person. strings a little bit, but we can make it happen. So it's actually really cool. We we were able to do a pretty significant digital transformation and modernize everything within that. Niel: Yeah, it's yeah it's another good point you bring up. I found early on I want to identify a really bad pain point when I take on a new role and solve that pain relatively quickly and easily. you know, if it's some type of of performance issue or process issue, whatever it is, solve that problem and you get tons of credibility and trust. That you can then leverage into doing all the other things that you know need to be done. So Felix: Yeah, yeah, ex exactly right. Getting getting approval to do some of those other things that we needed to do, but taking it from the business angle. Niel: Yeah. Yeah. Felix: how how's this gonna help business objectives instead of just throwing out techie talk that they they're not gonna care about? Niel: So your data two logistics, then where did you go next? Felix: I went to Academy Mortgage after that, and I was there for about five years. I I was initially brought in to manage the the network and telecommunications team. I know you spent some time there as well, but Niel: Yeah. Felix: about twenty five hundred employee mortgage company, one of the largest independent mortgage companies in the country at the time, headquartered in Draper, Utah, but we had somewhere around two hundred and fifty branch offices throughout. Correct. Niel: Yeah. 'Cause they'd grown through ac through acquiring other smaller independent mortgage brokers. Felix: Yeah. and then in in that five year span I I you know moved into a VP role and then eventually the CIO role. That that was my first experience at at the at the CIO. In fact, I sure recall we we hosted some type of event where where you came to to our offices and I I sure remember having a conversation with you Back in our in our WGU days, I remember you used to have these what do you wanna be when you grow up conversations. I thought you you'll recall those. Niel: Yes, huh, huh. Yes. Felix: and I still remember expressing my my desire to eventually be a CIO at some point. by the way, I I consider you one of my mentors, Niel and I you you gave me some great advice over time. Niel: Well, so so I only invite people on this podcast that that are going to praise Felix: Ha ha Niel: me. It's just a just a generalized rule. So Felix: No, you you gave me some great advice at the time, right? I as I mentioned earlier, I had come up through the infrastructure ranks. Niel: Yeah. Which is h it's hard to get into a CIO role from IT operations. It just is. Felix: Yeah, yeah. And and and I I remember you mentioning that, but challenging me and and actually putting me in touch with some resources, you know, here's some people that have gotten there. Let me put you in touch with them. And but part of it for me was getting a little more well-rounded around software engineering, data services, cybersecurity. so I actually went back and and finished a master's degree while while I was at at Academy Mortgage and actually focused on on those areas. Niel: Okay. Felix: Prior that, you had given me some opportunities to lead some software engineering projects when when we were at WGU data too, I had some opportunities to work very closely with them. so it it was actually really cool to to host this CIO group and be able to report to you, like Niel I I want you to be one of the first to know. You know, I I I've I've been appointed the CIO here and thanks to your guidance and a lot of people a along the way, Niel: So you did some interest I know you and I talked about this a lot. You did some i really interesting stuff at Academy with AI. And this Felix: Yes. Niel: is back in the days before Gen AI, when AI was machine learning. robotic process automation, some natural language processing, but natural language processing was its own challenge because anyway, talk to me about some of the things you did at Academy, because then I want to project that forward to some of the things you're doing at MX around AI, and maybe tie the some of the themes together. What does it take to be successful at AI independent of the forms of AI? Felix: so early on in in my CIO role, we saw a lot of opportunities to automate some of the processes. the mortgage industry has been around for forever and ever. and and frankly I I would say Niel: Nothing's changed since forever and ever. Felix: Yeah, nothing's changed, right? Yeah. te technology has not advanced very rapidly in that industry. But I would say over the pe past ten years, ten, twelve years, it it has certainly accelerated. But we had processors that the organization would hire to to do you know our The bread and butter is is our the loan officers, right? They they're out there processing loans. But between applying for a loan and and and that fun that loan funding, there's there's a lot of steps in between. So one of the first areas that we identified as opportunities for automating is was was the pre approval process. we had partnered with with with the company that handled that intake process of of filling out the form, right? So no longer did people need to come in to to the Office and and sit face to face with with the loan officer, they could do it all online. the timing was really interesting too. This was like late 2019, early 2020, and you know what was coming, right? So the timing around, hey, we need to be fully digital in what we do. Well, we got we were so fortunate that we had started a robotic process automation process. By the time the spring of 2020 hit, we already had A pre-approval bot. we we affectionately called it Amy, Academy Mortgage for You was the acronym. Niel: Okay. Felix: and this bot took about 23 steps of that pre-approval process that had historically been a very manual. I need to go do a background check, I need to run your credit, I need to do employment verification. and we were able to automate all of this with Amy using robotic process automation. So the minute someone applied for a loan, the loan officer would go in and just do some basic verification, and then they could either email or text Amy and give it some action item, start loan number, blah, blah, blah. And Amy would take it from there. Those 23 steps, they would just happen automatically Niel: how much time did that take days or weeks out of the process? Months. That's crazy. Crazy. Felix: months, honestly. Yeah. It it it was we we we we partnered with with an outside firm to to help build some of this while we built some in in-house expertise. but I would say maybe three months. but a lot of learning in between and you know, again early days of automation and some of our partners we we had to get creative and how we were interfacing with with with them. we Niel: That's still pretty fast for a fairly fairly comp that's a complex process. Felix: It was, yeah, yeah. A lot of API calls to different providers Niel: Yeah. Felix: and and whatnot. But at at the end of the day, like Amy would take 15 20 minutes at most to process a a loan file and then it would bundle it all up into this nice little SIP package that the loan officer would get an alert and say, you know, so and so's approval approval package is ready for you to go review. And so our our ability to go process loans a lot faster just increased dramatically and very fortuitous because Niel: Yeah, Felix: Unfortunately the the the pandemic happened, but Niel: COVID hit. Yeah. Felix: everyone had to go home. and we saw unprecedented volume. Like we saw volume on l anything we had seen prior to that because interest rates went down, we were buying homes. So Niel: Yeah. Yeah. Felix: the only process that needed to be done in person was title. Niel: Interesting. Felix: yeah, it ev everything else we were able to handle electronically. So Niel: Now I can't remember, did did you apply any AI to the underwriting process, like a machine learning model or anything like that? Okay. Okay. Okay. Felix: no, i it was primarily just RPA at at the time. eventually we we brought it all in house. we ended up choosing a different technology provider to to Niel: Okay. Felix: and we rewrote all of those processes in-house to where we we actually ended up embedding more machine learning and and that type of thing. But Niel: Okay. So, so what lessons did you learn about AI readiness and maturity? What did it take to be successful then? And is any different from what it takes to be successful with AI now? Felix: Yeah, I know y you hear this all the time and it's so true. Like your ability to deliver these types solutions it's only as good as the clin your your Niel: Okay. Felix: ability to to be diligent in in in in sanitizing your data. Niel: Yeah. Yeah. Felix: So data sanitation was such a crucial aspect of it. we we we had a really good data services team at Academy Mortgage that that was very diligent in in that regard. So it it wasn't super painful, but you know we we had to get our ducks in a row and like if we're we're we're gonna be throwing this RPA at at our data, it better be otherwise it's garbage in, garbage out. if if Niel: Yeah. Yeah. Felix: you're if you're not careful. so you know, fast forward to where we're at today with with MX. we were pretty early adopters of of AI. I like to think of AI kind of in three categories. you know, you have your baked-in AI solutions, right? It's it's your your we're a Google Workspace shop, you know, integrating Gemini into Google, the things you do with Google, you know, help me summarize, help me write an email, Zoom. right the transcripts for me those those those are the the baked in solutions those were easy for us to just literally go hit a enable button and and you know train our people and and and send some notifications then you have your standalone ai solutions your gemini's chat your claude your chat gpts and and and whatnot two years ago we were pretty early adopters in in getting Gemini chat out out to the masses and letting people start experimenting with it. July of this year we went all in with Claude, Anthropic Claude, co-work. We we signed Niel: Okay. Felix: a co-work, Claude Niel: Yeah. Felix: Work deal. But we saw an opportunity like, okay, what were some of the lessons learned when we turned people loose with Gemini? And we we had you know some training and whatnot, but for the most part people were expected to Go learn it, go figure out how to integrate it into your day-to-day work and and kind of left it at that. About a year and a half ago, our CEO sent a a company wide email challenging every single employee. Like, now it's your time to go learn AI, embrace it. It's it's the wave of the future, and and now it's your time. I we took that as a challenge within our department. I I actually asked every single individual within my department to go get two. AI certifications, their choosing. and everyone did. We we we actually went celebrated at Valter's after after that Niel: god. Felix: was done. but it was very fortuitous because we we then came back and said, Okay, we're gonna go all in with with Claude And I partnered with our learning and development team. so we've developed internal AI essentials course. we have bi-weekly Claude workshops. where we have different people within the organization come talk about how to build your second brain, how to get really good at prompting, how to build skills, things of that nature. And then I I think the best part that that came out of all this We took our our primary company objectives. I we for our fiscal year we have four core primary objectives. Every department is asked to build metrics around how what is your department gonna do to help move the needle on these four things. Instead of coming up with new metrics for AI specific work, we said no, how are you gonna leverage AI to help Niel: Yeah, right. Felix: towards those objectives? so when we talk about, you know. We we we built an internal token portal where we monitor utilization and things of that nature. we've even established a baseline of every employee is gonna get so many tokens per month. And if you need more, there's there's a short request form that you can submit that actually goes to the manager and and cost center owner depending on the amount. And and you have to provide some justification as to how are you going to use it and which of the company objectives is this helping so that the manager can have those conversations. So we we are able to track how those tokens are being and tell a better ROI story. Niel: Yeah, I really like that approach because your AI is not an objective. It's a tool you leverage to achieve the objectives. Felix: Exactly. Niel: And and that's how you can measure the impact. Because I think people are struggling with how to measure the impact of AI. To what extent is it helping me achieve the orb the organizational objectives and what 'cause that that applies universally generally as opposed to I have AI objectives. No, I'm gonna use AI to achieve my objectives. I think that puts it in the right place in context. So that's a great idea. Felix: Yeah, thank thank you. And and you know, we we think about it from from both an engineering perspective, the the product that we're building for our customers, and how is it helping us get product to market faster? but also at the enterprise level, how is it helping our finance team close the books faster? How's it helping our marketing team generate better leads? that that that type of thing. Niel: So what comes next with AI do thing, not just at MX, but in general? Felix: That's it's a great question. we've outlined five milestones internally that we feel ultimately we we want everyone to join us along this journey within the organization. So it starts from milestone one is just just learn the basics to milestone two is integrated into your daily workflow. Three is now where you start getting into what are some of the things that it can you can automate, tasks and and that type of thing. We believe strongly that that the the future is agentic solutions. Milestone four for us is start building agents. We often talk about you know, we we we talked about being peop the manager of people. We we strongly believe that every MX employee in the not too distant future will be managing agents. Manager of agents. in fact I I was reading just yesterday Gartner predicts that the average Fortune 500 company by 2028 will have somewhere around 150,000 agents deployed within their environment. And that's exciting and nerve-wracking at the same time. It is. Niel: That's a lot to manage, isn't it? That's a lot to manage, coordinate, govern. There's a lot of work to go into that. Felix: Yeah, but we we're we're fully on board. We believe that's that's where we're headed. Milestone five is and honestly we've seen small glimpses of it within pockets of our organization, but leveraging AI to create value driven solutions for the organization, whether it's product that we're selling to customers or internally generated. So we're we're barely scratching the service at milestone four currently, but that that's how we think about it. Niel: Good. Felix: I I was at a conference last week, Both Octane and Mercado hadn't had their same conference. Well, I guess it's two weeks ago. Okta announced the Blueprint Alliance is what they're calling it. They've partnered with about a dozen other technology companies, AWS, CrowdStrike, and a few others, to your point, around how they think about agents and the governance of agents. And it's all around answering four four basic questions, right? What are the agents that are out there? What have we Allowed them to do, what are they actually doing, and how are we going to respond? So we've got we're all trying to figure this all out together, right? We're all in this journey, but we're we're adhering to to this blueprint alliance Niel: Okay. Felix: white paper and and how we're thinking about agents, and and really they need to be thought of as just like we treat humans and giving them their own identity. That's the way agents need Niel: Yeah. Yeah. Felix: to be treated. Niel: Yeah, it's interesting. I had this conversation earlier this week. We were talking about AI coding platforms and the challenges with governance around those, when ever everybody can build whatever they want. And the point I was talking to two other CIOs, and we just realized if you look over the last four years. Of the challenges that came along with AI, we've solved them. So this AI code governance of AI coding, we'll probably solve that sometime in the next three to four months. Then we'll go on to the next issue and the next issue. But as they come up, we're figuring it out, which is kind of nice and hopeful because you think, how am I going to govern 150,000 agents in my earnings? By the time it gets there, we'll have figured it out. Felix: Agreed. You know, these are exciting times, like I said, ner ner nerve wracking but exciting at the Niel: Yes. Felix: same time because technology's coming at us so fast, but we're also able to solve a lot of these challenges a lot faster too. I I I was Niel: Yeah. Felix: just chatting with one of our software engineer VPs a couple of days ago. And and we're we're rediscovering the war the way work has been done all along, like how long has Agile and Scrum been around? We're Niel: The the the Agile Manifesto was 2001, so we're the twenty-five year anniversary of it. Felix: So we're seriously with in MX, we're we're challenging that status quo. Is is scrum still the right way for us to deliver? Niel: Yes, I agree. Cause it might be it might not work in the new environment. The way the well it Felix: Yeah. And what Niel: might not work in its current form in the new environment. I think the principles Felix: Exactly. Niel: hold, but the application of principles will be very different. Felix: Yeah, for sure. So what we're seeing, again, part part of the conversation I had with with this individual, we're seeing from a code production perspective, with the same number of people, we're producing about twenty percent more code. Niel: Yep. Felix: with with with the help of AI However, that bottleneck is moving downstream. So no Niel: Exactly. Felix: now it's no longer we're not developing fast enough. Now it's integrations, testing environments, deployments, Niel: Yep. Exactly. Felix: those are the things that are now the bottleneck. And and we'll go figure that out. Like Niel: Yeah, we will. Felix: it it it's interesting how that just you know the ball kind of moves down the the road. Niel: Yeah, it's interesting. I I've talked to a few people about this. we've built entire methodologies around what was the truth, that it took a lot of time and effort to write code. Well, if it no longer takes that much time and effort to to write code, where how do our methodologies need to change in an environment that assumed that that was going to be the bottleneck? Now, like you said, the bottleneck somewhere else. So we got to change the way we do things. It's it's an exciting time. And I'm glad Felix: It is. Niel: I'm old, so I don't have to deal with it other than to, you know, critique. advice. Felix: Well the the the the neck the cool thing is at MX we have seven core values th that that that we live by. One of them is purposeful contribution, the other one is iterative innovation. And when when we think of it from that perspective, we are constantly challenging the way we do things and redefining so I I'm glad to work for an organization that supports it and and expects Niel: Yeah. Felix: it out of us. Niel: Well, we've spent some time with Felix, and I hope you had enjoyed as much as I have. There's a lot to learn. Felix's transition from IT operations from a technical leader into a people leader, and what he learned along the way, and what that implied about his role and his behaviors. And secondly, an opportunity to learn from somebody who is doing AI. pre-Chat GPT. and how to extend those because what I found find in talking to people like with peop with people like Felix is others are learning those lessons that Felix learned some years ago and those lessons still apply. And so it's good to listen to people like Felix that talk about data quality, that talk about governance, all these things, because those have been a reality that some people are now having to do the crash course on. So anyway, I appreciate your time, Felix. You're a good friend. I've appreciated watching your career develop and all the good times we've had getting to know each other. So thank you again, Felix. Felix: Thank you, Neil. really appreciate the time. always a pleasure chatting with you. these are exciting times that we're in and and I'm glad that we're we're right in the thick of it trying to figure it all together. So appreciate the Niel: Yep. Felix: time. Niel: Thanks.