LordsOfComputing012 - Brian Gregory at ESI

CotŽ: [00:00:00] We'll go again it's another exciting lords of computing podcast with me CotŽ and my co-host when I say hello. 

Matt Curry: [00:00:06] Hello everybody. 

CotŽ: [00:00:08] How's it been going for you , Matt. 

Matt Curry: [00:00:11] It's been going good. Yeah. Still hanging in there. 

Matt Curry: [00:00:14] Few changes hang in there. Well making progress. Continuing the fight the good fight. 

Matt Curry: [00:00:20] And yeah it's just new new things to learn and new things. You know to go tackle every day. It keeps it interesting keeps it fun. 

Matt Curry: [00:00:32] You're such a modest person I'm going to promote something for you. Before we introduce our guest I think you've secured a speaking slot at the Cloud Foundry Summit this year right. 

Matt Curry: [00:00:43] I did. 

Matt Curry: [00:00:44] I'll be talking about how we've built a brand around our enterprise digital transformation. And you know how that plays into organizational behavior and some of the benefits we've seen from that and why I think it's an important step towards you know driving your overall transformation. 

CotŽ: [00:01:07] It's like the whole theme of this podcast almost. 

Matt Curry: [00:01:11] It is boiled down into like 20 minutes plus 10 minutes of Q&A. 

CotŽ: [00:01:18] Well hopefully. I'm pretty sure they'll be recording. We'll have to track down their recording but also if anyone wants to go to the Cloud Foundry summit they gave me a discount code of C-f 16 cotŽ. If you go look at the Cloud Foundry someone will put this in the show notes at Lords of computing dotcom but you can get 20 percent off there. But yeah definitely maybe I'll even steal that recording and stick it in here. Or no. Steele is it technical or a metaphoric term but maybe I'll try to put it in this area. That should be an exciting talk. So anyways we have a we have a as usual a guest episode. I was introduced to him by one of my friends at pivotal and he's been he was brought in I think a few years ago to help do transformation basically at the company he's working at. And we had a little a little call to get to know each other and he seemed perfect for shooting the crap here with Matt and us on the podcast. But why don't you introduce yourself. 

Brian Gregory: [00:02:17] We'll do things cotŽ. So my name is Brian Gregory. I'm with Express Scripts. 

[00:02:23] I joined the company a little over two years ago and I was brought in to lead the database organization. It's a pretty sizable org. It was just shy of 100 people. Previous experience was in the cloud world where at Savvis which century when all got that good stuff. So I was kind of exposed to the you know transformation journey I guess years before we were trying to invent what cloud was working in that space. And when I came here it was supposed to go down to database space and I thought well that's kind of interesting you know I'll just be able to have one focus and figure out where that goes. 

CotŽ: [00:03:01] Right you're like finally I can do just yeah. 

Brian Gregory: [00:03:03] This will be right because before I had you know the VM Ware team I had the days I had network I had all kinds of fun stuff. Now I was going to focus on just one. When I came to express scripts you know that was my focus after about a year over year we started having conversations about transforming the company and really been on a bender of M&A for years. And that lent itself nicely to a you know agile development transforming. 

Brian Gregory: [00:03:30] It was more about isolating to one platform and and getting you know. 

Brian Gregory: [00:03:35] On board in the New. New companies. So. 

Brian Gregory: [00:03:39] They decided they were going to take this take this on and decide they're going to transform. I started becoming part of those discussions just because of my experience of knowledge. Soon enough they knew me that full time and I'm now leading the cloud strategy and engineering team. So I still have the database team there actually back. So tell me if you're interested in the plan for that. Well you know it's like like we talked about before I go here's a promotion go take this on and you know I went from a 100 person organization to just me I didn't have an org. 

[00:04:13] Oh that kind of promotion. It's like what is going on. 

Brian Gregory: [00:04:19] That's the funny part about that is that a lot of you know leaders BP and whatnot said Are you sure you're OK with this. You know like you're you're leaving this person or for an unknown if you will. And sometimes I say things I probably shouldn't but I basically said that the good thing about doing this is if we can't figure this out we can't transform that. I know that I want to be here anyway so I'm OK with taking it on like I'm not concerned that you know it's going to fail in three months and you know we're going to regress and go a different direction. I feel like the company was at a point where we were going to transform No we're willing to. 

Brian Gregory: [00:04:57] Take on and do that. So that's all. 

Matt Curry: [00:05:01] I also struggle with separating my inner and outer voices sometimes. So you know we have that in common. 

Brian Gregory: [00:05:07] Well I know from our conversation Matt that we do have that that is you can speak from the heart and I'm not sure that's always a good thing. But but it's an honest thing. 

CotŽ: [00:05:18] I have this big company anti-pattern which can also be a pattern if it works out well called asking questions usually results and assigning yourself homework. 

Brian Gregory: [00:05:28] Right. T. 

CotŽ: [00:05:30] He more passionate involved you are you know again this can be great. It's sort of like if you if you care and you want to get involved and you make that known Hopefully you're in a culture such that people who care and want to get involved are given work. I mean that's why they call it work. It's not called you know hanging out but you do have to be a little leery every now and then. Oh crap I just created a bunch of word for myself. 

CotŽ: [00:05:52] If that's not what you're investing. 

Matt Curry: [00:05:54] In the enterprise when you ask questions that people don't like they elect you to a committee. That's how they get the job done. 

Brian Gregory: [00:06:02] You know one thing is actually kind of funny about that is that when you know our leaders basically made the investment and said This is how we're going to springboard into this journey. When I was asked about that kind of review time and they said you know where do you want to be ADSI And what do you want to do. I think the obvious answer was to go back to my roots and get back in this cloud you know space and the the abstains and I just basically said I think I'm good. There's a lot of work to be done the database space. You know I'm OK here. And you know a week goes by and they asked again are you where do you want to be and what do you want to do and then my leader at the time said you know. All right well how could yes this use you know going forward which was a different question. I think my resistance a little bit was to say how serious are we all still kind of in this mode of like. 

Brian Gregory: [00:06:50] I'm seeing a lot of things but I'm not you know they're all words of lip service and I like it as you know you guys know [00:06:55] it takes a lot to transform an organization and you need that leadership you know top down approach. [5.8] And I did start seeing that and realizing that we are we are in this and that's when I said OK here I come. 

CotŽ: [00:07:08] So yeah I think I think that's a good point. That's sort of like the the key question to ask when you're about to assign yourself homework is am I going to get the support that I need. And then a broader question is do they really mean it like that. Is this just some canard that I'm going off on or is this it's actually mission critical to use to use an old funky term for it. And it's important that this stuff is sold write is it. 

Brian Gregory: [00:07:35] Are we doing it just to be relevant or are we just used to say we're doing this and you know are we going to really transform our business. And so that's when I was convinced that we are serious about it. Of course at that time I didn't have any resources. You know we're getting there I'm certain to build an organization and I've got a handful of people now. So actually Matt and I talked about that before about what what does that look like and how can I. How should I approach this because [00:08:02] it's not one of those things that is going to go ramp up and say I'm going to build an over 20 people. Nor can you run around say I'm a stand up dev ops organization [9.7] which I think is funny in itself. You know this is like it's not something you can build and in preparation for something it's something that you know build as a result of something else and it's kind of a culture thing. 

CotŽ: [00:08:25] So yeah that's a good a good segue to like kind of the next topic which is what will tell me what you both both of you all think of this like. I find myself so in that situation you are talking we don't just go like build an organization right off the bat or whatever scurrying behind that is this I don't know I would almost call it a theory that when it comes to I.T. transformation at least the way that we're currently doing I.T. transformation whether you want to call it like cloud-native or DevOps or cloud or whatever the first thing that you have to realize is how we end up doing this. We don't exactly know right now. So the first step is to figure out what we're doing and forget till we figure out what we're doing and why we can't really get to the how of it how we're going to do it. And then once we get to the how of how we're going to do it then we can build an organization. But the trap you don't want to fall into is like well of course we're going to need 50 people and a data center. So let's build those first while we're figuring out what it is we're doing. 

Brian Gregory: [00:09:30] Right. 

Matt Curry: [00:09:32] I don't know. 

Matt Curry: [00:09:32] I mean I think like to me you kind of have to iterate into it because things will you just don't have all the information up front like there is no [00:09:41] you can't take a waterfall approach to like building an enterprise transformation like you have to be comfortable with living in this unknown area. [8.3] So if you think you need 20 Well maybe you know you need five start there and like see what you can build with five and like see if you can build some momentum and credibility get some learnings get customer feedback and operate from there. I think the other thing that's important is we talked about leadership by and and yeah it's important it's hugely critical and we're all sitting here saying OK our leaders really are 100 percent right and behind this are they showing us that reactions. But the other thing that's important to understand is [00:10:18] most transformations don't start with like CEO level buy-in like you don't get to start there like you have to get there. You have to win yourself there to get to that finish line but it's definitely a milestone that needs to be crossed at some point right. [20.1] 

Brian Gregory: [00:10:39] Yeah that's a good point. That's a good point and [00:10:42] it's not just the at that level the sea level there's a lot of other business folks that you know you have to win over as well. [7.5] 

Brian Gregory: [00:10:50] Like you're talking about. 

CotŽ: [00:10:51] So when you were sitting there as a transformation of one how how how how did you start going through this. I don't know for lack of a better phrase sort of enterprise strategy discovery phase where you are like building out and trying to figure out what to do and. And I'm almost almost right before I don't know maybe not maybe I'd characterize it differently but right before you start iterating on an idea like what do you how do you bootstrap this from from from one person. 

Brian Gregory: [00:11:20] Well you know which was kind of funny is how we really launched this was a little bit backwards I think. 

Brian Gregory: [00:11:25] We were trying the approach of OK now we've bought a you know a converged infrastructure solution. We're going to we decided to go PCF. And we were talking about having these dojos and we try to go with this Qumbyyah thing where every one condo was included and part of it and I think you had the traditional I.T. dragging its feet or you'll never come to a consensus. So nothing could happen. Me and my open mouth you know like Matt I went to my boss and said kind of unleash me let me go after this and I'll I'll get this thing going and we'll will kick off. 

Brian Gregory: [00:12:03] And I finally got that permission and I basically set the meeting up at the first dojo was like five different business partners. 

Brian Gregory: [00:12:11] That I brought in that had what I thought were viable applications that would make sense to have this kind of conversation about what would it mean to rewrite or app or and over or what can we do there just to get going. [00:12:23] I knew that it wouldn't be the right. 

Brian Gregory: [00:12:26] End state but it was something to get us going and start to integrate on on something right. [8.2] And so you know example of that as that guy came in and said editing the CMS project and so we started out you know with Drupal and then we went to another solution and we enter five times and now he's where he needs to be. But that. That was never going to happen and what we just said let's do it was [00:12:49] let's make the call let's go in the room and start kind of walk in because he needed that kind of knocked down a bunch of barriers and get the ball rolling on that one show never you know defined that end state it's not going to happen nor should you. [11.9] 

CotŽ: [00:13:02] So it sounds like the time if I get this wrong that's the sort of timeline and then tactics he got into was you started off more or less with like a centralized like here is going to be our cloud or whatever our new way of doing things. And then and then and then you kind of detected that there was enough like resistance or as you said foot dragging and you're like This isn't really going to work out as quickly or as well as we wanted it to. 

CotŽ: [00:13:28] And then you sort of shifted to and this is my characterization of it. [00:13:32] I need to go out into the organization and basically find my customers and I might need to actually tell them they're my customers and they may not realize it but they do effectively market that we could be doing something for them right like they should instead of doing this all on their own they should think about using the sort of cloud thing that we're setting up. [21.2] 

Brian Gregory: [00:13:54] Correct. That was absolutely [00:13:56] being the Evangelist which I didn't realize was going to be my primary role was was really when it came to me. [6.7] 

CotŽ: [00:14:04] Yeah. And that's interesting because it is a I mean I mean you know I think I think you've had a similar experience right Matt where you. You need to kind of pull some people in to the field of dreams you might dream build every now and then. 

Matt Curry: [00:14:19] Yeah. That's exactly right. [00:14:23] I was you know we took the approach of really just getting together with different areas and seeing where we could find excitement or people who had experienced so much pain dealing with traditional you know infrastructure land that they were really ready to try just about anything else which as it happens isn't that hard to find those people. There are a lot of people frustrated especially on the application side by a traditional infrastructure land. [34.4] And I think finding those business partners and getting some early wins is hugely critical. So I was wondering how your database experience played into exposing who the who you could partner with or if it helped you at all having come have having had a history in that space because you know persisted and this is kind of one of the harder parts of distributed systems. 

Brian Gregory: [00:15:28] Yeah, it's um... I would send the relationships where we're there because of a lot of the teams you know consuming database services and. It helped a little bit when it comes to that. But I don't know. Our traditional abstract I wouldn't say it was not like the the low hanging fruit if you will. They were. When you start talking to those teams it was kind of like how did they play. And that's that's a whole nother discussion about the cloud native roadshows I was telling you about. But it's it wasn't them as more the. What's the Greenfield stuff. And of course you don't have that many of those so we had to go after those. But then it was going back to people that I've got relationships with and have partnered with in the past. And oddly enough some some of the stuff that came out of that was pretty interesting because you know they were doing mainframe COBOL. 

Brian Gregory: [00:16:20] Stuff that would never even go in a sentence with a PC. And we found that we started doing other things. [00:16:28] We had a hackathon our very first hackathon with that same group. And I think that was basically because on relationships and talking about what we could do that would be kind of different and cool and deliver you know business value. [12.7] So I think that really helped. 

Brian Gregory: [00:16:43] But I think at the end of the day the database space thing that scares me is I know too much about it so I know that the complexity of that database is a service and how are we getting things going forward. 

Brian Gregory: [00:16:54] Right. Moving things off of certain technologies expose them via API eyes putting them in maybe a cached solution. You know theres a lot of ways to skin that one and I dont know that were going to be able to go fast enough because in my analogies it's like you break down the stairs you know as layer and it's like OK we can automate that and then we can do a lot of PAHs But then. The 

Brian Gregory: [00:17:17] next you know the next day and you get to do that database because that's where the data is at. And if everybody's has their touch points there you're going to have to move that data around and make it exposed a little bit better than it is today. 

CotŽ: [00:17:31] It does seem like the what would you call it maybe the second struggle. In doing a cloud native transformation like the first one we kind of alluded to is just like getting people to want to make create and run their applications in a new way. Technology and process wise. Right like right just getting them to kind of like an agile small batch DevOps-y kind of thinking and right and then you sort of move people over there and then it seems like you know you kind of like around the next bend and you're like oh crap data. And then that's like that's like it's like a whole new set of problems that you have to solve that you know maybe it's because I don't really follow data too much because I gleefully kind of exist in the stateless world more than state but it doesn't seem like there's like a whole lot of solved problems there. Like there's there's a lot more questions than there are answers and so you have to go in there and figure it out somewhat painfully. 

Matt Curry: [00:18:32] I think everybody does it slightly different. And so I think that's why we haven't seen like a really massive database as a service offering outside of you know what you see in a U.S. raid. So it's kind of an interesting dynamic there. One thing that we've seen on our side is you know for staging and like testing environments we just gave really simple inadequate solutions at least to allow developers to be unblocked and as long as we provided them with the information that it's not the final answer. They were good with it because they were good enough with it that they could at least move forward in like persist data somewhere. But knowing that they were going to have to change have a day to go I persisted in the future. Enable them to write their absence such a way that you know they knew that change was coming and that change would be easier. I don't know what Brian's seen on his side but that's not. 

Brian Gregory: [00:19:32] You're absolutely right. I was just going to say it's one of those things that we we solve for the interim but we haven't. 

Brian Gregory: [00:19:39] We know that's not the long term solution that it might change. And we're doing things. 

Brian Gregory: [00:19:44] The one thing that's interesting too is that you know you take your traditional database space and they they typically don't support my sequel rattus you know Mongo things of that nature so bring in that conversation and mix is going to be a whole another aspect as well I don't know what you guys are doing there for you. 

Brian Gregory: [00:20:04] How do you have that managed or what not today but that's going to be a bit of a... Now I will say that the teams interested you know that the team when I say the team's traditional stock you know it's Microsoft sequel Oracle and Teradata you know DB2 to your typical database platforms. 

Brian Gregory: [00:20:24] When you start going to the open source community and things like that it's a little bit different for support. 

Matt Curry: [00:20:32] Yeah I think in our world that Oracle is the answer to every problem. 

Matt Curry: [00:20:37] And it's just unfortunate but we're seeing exactly the same thing where we're starting to see enough momentum that we need enterprise offerings around Mungo and some other some other different type of data stores. You know we're trying to kind of ground the architecture in cap theorem so that you know that is where the distributed systems are architected aligns with how the persistence is architected and trying to send that message. But even that is a pretty big shift for how we've typically approached data. And so I guess we have a long hard road ahead but an exciting one. 

Brian Gregory: [00:21:23] Yeah. It's like the The funny thing is that's like the three year map or something you know where we're obviously both laser focused on the you know the next week while we try not to look too far out. 

Brian Gregory: [00:21:36] I don't speak you and I that is we try not to look too far out because it's like we've got so many prompt right there in front of you that you kind of have to address. I mean it's still win and over you know I think. The interesting thing is that I'm starting to feel that moment. I'm so lucky you create this way because you want people to believe in what you're doing and consume it and come to your you know come to your new home and watch their awesomeness. 

CotŽ: [00:22:02] I mean you don't want to crying alone at your party. 

Brian Gregory: [00:22:05] Right. 

CotŽ: [00:22:05] You want to be crying with people. 

Brian Gregory: [00:22:07] But at the same time it scares you because you realize you've done that you've done just that. 

Brian Gregory: [00:22:11] But you don't have the back you know the back in forces behind you to support that. So that's when you start to scramble you know day one I'm like you know depend on what random articles you read. 

Brian Gregory: [00:22:23] If you talk about days year on day one staffing You know it could be five people it could be 50. 

Brian Gregory: [00:22:31] I was more about let's just kind of see where we go and see how this takes off and then you know go from there. Well now I've done that. And so now it's like people want to go to production in the next month we'll probably have three apps and pride. So now you're scrambling going on. What about the nonfunctional requirements let's make sure we get all that stuff you know nailed down and we are talking about some you know pretty heavily used production services. 

Brian Gregory: [00:22:54] So it's not just fun ones on the back end that I can play with and show off to my friends and say look at this cool app we get to that point. 

CotŽ: [00:23:03] It's always good in the middle of a conversation like this to actually define what your organization does. 

CotŽ: [00:23:08] But since we've gone over kind of characterizing the sort of applications and services. 

CotŽ: [00:23:16] Can you give us kind of an overview of as as as you so easily say what each side does and and how that kind of decomposes into the type of I.T. that you have to have in the applications that you need to run. 

Brian Gregory: [00:23:29] Yeah I mean we're you know it's funny because where you know P.B.M. in business but you know our goal is no different than what it is for like IT So our user experience is what we're after. 

Brian Gregory: [00:23:40] Right. It's the whole world today is delivering a better more affordable user experience to our end users our customers you know whether that be. 

Brian Gregory: [00:23:50] You know you go on online and try to figure out how you get your mail order scription we want that to be a seamless integration process for people. So that comes into the digital side right then you get the Web into Mobile and things like that. So we're trying to go after creating that user experience that you know let's be honest a lot of people have nailed in the industry not naming names but we're going after that. Is no different than what we're trying to do with I.T.. Right we want we want to be the PaaS in the IaaS layer that says it's here it's available Here's all the capabilities we're going to and empower you to be awesome and develop code and write code that basically will turn into revenue for the company or you know a better experience for our end users or. That's kind of what we're trying to focus on at this point. I think that you know I think that the the big thing is for us it's hard because you know we are you know Fortune 22 company right we're big you know a billion dollar company so it's not that easy to just course correct the Titanic. So we're trying to figure out where those opportunities are and we can you know turn around and make things better you know without disrupting the backend services and all the things that run our hundred billion our business. 

CotŽ: [00:25:06] Right. 

Brian Gregory: [00:25:07] That's a difficult part right because you can't just forget about that that is what your revenue is. You can't just go off and do all kinds of fun things to disrupt that. 

CotŽ: [00:25:15] Yeah and this I mean this is always a can be a bit of a rat hole or a rabbit hole. 

CotŽ: [00:25:21] I don't know which one is better but like how do you in that context how do you sort of determine what better is when it comes to software and end to kind of explain that I was. 

CotŽ: [00:25:35] I think I was reading through some Adrian Cockcroft piece the other day and he pointed out that like a lot of this like cloud and dev ops and stuff we talk about nowadays it matters a lot if you care about the user experience and the design of your application because you know if you're if you're going to be like releasing applications every day or every week in theory in addition to be able to patch things really quickly. But the reason you're doing that is because you're paying attention to user experience or design and you're trying to make the interaction between a human and your software better. Which right. You know that sounds awesome but we've all worked in various big companies and I'm sure our prison employees are not like this. That's not always the corporate priority right. 

Brian Gregory: [00:26:21] Right. 

CotŽ: [00:26:21] Like there are other things that can define what better is. And so I'm I'm curious when you're looking across that and the initiatives that you're doing are very much. 

CotŽ: [00:26:31] I mean the the the reason you would use something like a Pivotal Cloud Foundry or things like that is you know largely about being faster or deploying and therefore I would think about better user experience but there must be other considerations as well or maybe that's the only one that defines what better. 

Brian Gregory: [00:26:46] No that's that's actually a perfect analogy because you know my my thoughts are you know [00:26:52] we process like a 1 billion prescriptions each year for tens of millions of patients. [5.8] I don't know that the patients really care about how. 

Brian Gregory: [00:27:01] Claims are adjudicated or the back end processing. 

Brian Gregory: [00:27:05] I think the things that they're exposed to is what we've got to nail down. Right. And the typical I.T. you know here six months years or IBM here start to write code a year later and you realize that you are way off mark and the customer is not happy and they're pissed and they're going in different directions. This is about. Any user interface right. So if they're touching they're doing anything you know maybe they're not touching it maybe it's on the back end it's faster it's seamless. If they're doing a pre-auth check and it goes quickly and they're not standing there in a counter somewhere or waiting for their prescription and mail you know things like that is really where we want to focus so. 

Brian Gregory: [00:27:44] We have been kind of focused on the B2C web and mobile app. Because that allows us to iterate and change you know daily get that automated pipeline and do things faster. [11.0] And that's really what this is all about. 

Brian Gregory: [00:27:58] That's not to say that we don't go after some bigger chunks down the road. I just don't think that they're you know in our in our near-term vision it's again if you're processing scripts on the back and adjudicating claims that doesn't sound like a big exciting thing at least for us three it doesn't right. 

Matt Curry: [00:28:14] Yeah well I think it's this is a super interesting topic to me because [00:28:20] I think it's actually a mindset thing like the people that are building your mobile apps and your web apps today are much more likely to be of the mindset to enable you for success to move towards a self-service platform versus the person running like your back and system that's written in COBOL and it hasn't been touched for 30 years [20.6] who probably is like oh this cloud thing is just a regurgitation of something we were doing 30 years ago and like you got you kids don't know what you're doing you silly geese. 

Matt Curry: [00:28:52] I think the core systems the core systems conversation is just like this, like this big lie. 

Matt Curry: [00:28:59] It's really weird like I'm trying to figure out where it came from because that's the way I think about it it's like my wife watches all these crazy HGTV television shows where they do like these renovations and you know every once in a while somebody goes and buys this really old house and as long as the walls are up like everything seems really great. And then like once you start ripping into them you start realizing that like there's electrical hazards and like the plumbing is done all wrong and there's these horrible things that have occurred over the years these barnacles that attach themselves to the ship of this court system and [00:29:36] the risk is actually exposing that those things are they're not like. [5.8] 

Matt Curry: [00:29:44] Not raking stuff yeah it's fun I think. 

CotŽ: [00:29:48] I think if you have a night or even a modicum of a program or background you can detect traces of this in various systems that use like well I guess to name names like you look at American Airlines log in and instead of putting in only your like username and password you also have to put your last name in there and I don't know what that means but there's some kind of like there's something kind of weird thing going on in their core system probably. This is when they merged with U.S. Airways. And you know it works. You know American's Web site is actually pretty good for what I would expect. 

CotŽ: [00:30:23] So kudos to that. But it is you see these little things pop up every now and then that are strange or like you know it's almost the equivalent to be very non-ID. I always notice when I go to visit my doctor or a clinic like everyone is constantly asking me for my phone number and my name like every time I see a nurse or a doctor and it's I guess it's almost like some sort of checksum in the system but there are similar things that you see in big enterprise systems where there's obviously some weird COBOL or as you say or something running around in the background that's precluding a nice or user interface like the abse it leaks all the way up to the top which which can be annoying. 

Brian Gregory: [00:31:00] Yeah. Yeah I think it's funny you say that because I think sometimes I find myself reiterating that you know our goal at the end of the day is to provide a platform and that's going to transform how we build software. Right. And people. Tend to want to take legacy processes which you know are kind of anchors. So every time you touch back you know who a one of my buddies told me that it was like [00:31:25] every time you touch back into legacy processes you attach another anchor which eventually will weigh you down and sink your ship. [5.6] 

CotŽ: [00:31:31] Right. Just like Matt's house renovation problem. 

Brian Gregory: [00:31:35] Right right. It's like you know it's funny. You know I do I am kind of cheesy right so I will I will use the dojo. You know what I do throw out the there is no fear in this dojo and all that fun stuff. But you know it's like I like to say it just takes smart people right. Partner with other smart people on the product side. And what has enabled them come up with cool like is and iterate until we come up with something of value and that's really what the focus is. There is no titles in the room right. They all drop of the door. We come in and [00:32:03] we're all focused on outcomes and goals [1.8] and not not focused on. 

Brian Gregory: [00:32:08] Reintroducing you know the old legacy processes or whatever let's figure out how do we make this. 

Brian Gregory: [00:32:14] Go better faster and do it in scale. Because you know that that's really where the winds are. 

CotŽ: [00:32:20] Yeah. And that touches back on something we were talking at the beginning is the. 

[00:32:24] The. The mentality that I find people often have to get over in situations like these is I don't know to put it in hokey terms like you have to tell your team like we need to go be creative and explore things like yeah we're not. Cause most of the mentality I think people have and. I don't want to say regular jobs but I don't know what else to say regular jobs is. It's pretty clear what I'm supposed to do. Am I'm in a nice way that's relaxing and beneficial and profitable. I'm a cog in a machine and so I do these things whereas in a situation we're trying to improve something or transform it's almost the opposite. It's like I don't even know if you're a cog. You need to figure out what we need to figure out what's going. Like we know what you're saying we know what outcome we want right. Like right we and we kind of don't know how to define that outcome but we know that people should be happier that we process these scripts and claims better or whatever. But we need to figure out how to do that. 

Brian Gregory: [00:33:21] So we're going in that direction and that's what we know. 

CotŽ: [00:33:24] Exactly. Exactly. And it's not really that that related to what we were just talking about. But one other thing I was I was interested in saying maybe this will be the the last thing before we wrap up. I knew we would have to do like a two part episode or something would be fantastic. But. So so thinking back on and you know at one point there was like you as the transformation person and then you had maybe like let's say three or five other people helping you out and you kind of had identified let's say two to five other applications you know customers who we're talking about earlier that would be working in your orbit here like you know hindsight's 20 20 and everything but looking back on it what are like the first two types of people or roles or staff that you think were handy or would have been handy do acquire like would it have been nice to have a project manager or like developers or. I don't know like coaches or something like the outside of your own skin. Like what kind of people would you staff a team with initially for an initiative like this. 

Brian Gregory: [00:34:28] I think you know your first and foremost is that I you know I always call them Swiss Army knife that are kind of. 

Brian Gregory: [00:34:37] The guys or girls that you could throw them in anything you know that they'll figure it out and be successful and and you know they're very very heavy on the technology side I think and and trying to understand things and the way their brain works. 

Brian Gregory: [00:34:51] I think those people they get so excited about learning something new that they don't hang on to existing processes. I think you get people that have too much process and they're going to be kind of going on the wrong tread and you know track and wondering well am I doing the right thing or wait. 

Brian Gregory: [00:35:08] How do we do this today and not trying to think of things different. I think the PM is critical at this stage meaning this age of three months and because you start realizing that you drummed up you know this way that's coming and you've got a lot of apps that are now asking to about pride deploys and what that looks like and you start out on the track that cause I'm not one to I'm not a very good project manager. So I think drumming up the business being that you know getting out and getting in front of having someone there available to you. As far as teach them how to fish right get them in the foundation teach them go through a couple examples do a couple of push deploys and once they realized that I think the right people kind of just take off with it on their own. So you know some won't but that's OK. They'll they'll need more education and we'll get that to own. But some. 

Brian Gregory: [00:35:59] They're just pretty hardcore programmers right. 

Brian Gregory: [00:36:02] You put something like that in front of them they'll start to realize the opportunity is a matter of fact. One of the funniest stories we have is a guy started out in the morning. This is a true story by the way I can actually show you prove it. He said I'm in the phone. I'm in the foundation now. 

Brian Gregory: [00:36:19] I need a Tomcat farm. I don't see a tomcat varmints like I. So we explained the whole Tomcat our thoughts on that later. 

Brian Gregory: [00:36:28] He was trying to push the command that is about noon and he's going I'm trying this command I'm not clear. It's not pushing the app and well we corrected him on a few of his commands. By the end of today the end of today he sent an e-mail. This is true. This is the true story. It said prepare the cloud for production. It's like I am ready I am ready to release to prod. It's like OK. So this morning you thought you had a you know the Tomcat you needed you know. So my point is some people now again we don't let that guy go to a crowd like there's a whole other process for that. 

Brian Gregory: [00:36:59] But some people will get in there and those Darco no question or happen it's you know it's great. It works fast and they had 20 different things in the pipeline that you need to automate to get them to prod and some only need five because they don't have touch points into existing databases or anything like that. It's a little faster. 

Brian Gregory: [00:37:17] I think the key is getting some people that are Swiss Army knife that can sit with them and try to figure out problems because really what you need is people that are going to try to sit down and say OK I understand your problem, how do we want to approach it? It doesn't have to be the right answer. It doesn't have to be the end answer has got to be a answer. You just can't say we don't do that or you know that's not how we do things here [22.4] or you know kind of kick them away and I think you know those people in your in your you know groups is they're kind of the I would say the rogue. A little wild. 

Brian Gregory: [00:37:50] Wild Wild West they're just more about solving problems as opposed to process and procedures. 

CotŽ: [00:37:55] Right. Right. No they're I mean to characterize them in a sardonic kind of way they're they're they're like problem solving junkies. 

Matt Curry: [00:38:06] Right. 

CotŽ: [00:38:06] Likes it like all they want to do is solve problems which. And this isn't always a black and white thing but like if you're only a problem solving junkie it means you're terrible at long term support for an effort and maintenance for an application. Right. Because you're kind of like there's no problems to solve here it's very boring but they're great for like that kind of this initial. 

CotŽ: [00:38:26] As you say Wild West of stuff. 

Brian Gregory: [00:38:29] Yeah exactly. And I think. 

Matt Curry: [00:38:31] There. They're your GSD cowboys like if you go to anybody in your Oregon like hey I want to do something weird. Who will help me get it done like. 

Matt Curry: [00:38:39] You'll get the saying no. Three to five names. Every time. 

Brian Gregory: [00:38:44] Right. 

Matt Curry: [00:38:45] Those are the people. 

Brian Gregory: [00:38:47] We have been successful. You know the rest of the company you know the different groups have started to help support this initiative. You know we had you know our enterprise application management team put forward 10 people to get to interview for a potential spot on the team. You know they're putting their best people up you know their their best kind of their lynchpins and their org which I think's a big deal when they're they're willing to kind of Donate. If you will to do the cause and know that they're going to lose this resource. I think that's a big deal. But I should say that while that's all important you also have to start to make sure you have those people that are very process driven to help guide you down the path for production support and the nonfunctional requirements because otherwise all of us are kind of wild and crazy people come up with all these neat ideas and then they're not they're not ever going to get to production because we don't have the other side to augment that effort. 

CotŽ: [00:39:43] And to that point I mean again tell me if if your experience isn't like this but I would assume you have, as Matt calls them, your I think GSD stands for get stuff done. 

CotŽ: [00:39:52] If I remember right that's right you have a You your GSD cowboys or your TV like police or whatever and I assume you know because there's there's kind of a. Process if you will that you described which is you need to establish some early success which then you kind of take to the mainstream and then that wins them over to doing things the new way. So I assume you get your Swiss Army Knife people and they're maybe not exactly the way you'll be staffing and operating the mainstream projects but they prove out this way of doing things. 

CotŽ: [00:40:25] And then if that goes well you can kind of take that to the rest of the organization who may not be Swiss Army knives maybe they're just really stable chefe knives like just nice functional things you can depend on and it's easier to win them over because you've built up basically as we would say in vendor land some cases some references that you can prove out to other people to win them over. 

Brian Gregory: [00:40:48] Yeah. 

Brian Gregory: [00:40:49] And one last thing I think that is kind of was a big deal for me which is going to kind of sounds silly but you know we're talking a getting introduced to Matt and then yourself because you're beating your head against a wall every day trying to again redirector the Titanic if you will. And then when you talk to people like Matt it's like OK he's been there. Same experience. This is not you know I'm not crazy this is kind of normal. And here's kind of kind of help. 

[00:41:15] Guide you down the river right you know where you want to end up but maybe you take a turn here take a turn there and once you know where do you need to take stick your foot in the ground and say I can't budge on this but I can on this you know that stuff helps because if you do start to feel like you're going crazy and I think I still quote quote Matt as lifting my spirits one day because he basically said hey ask me how I thought I was doing and of course I'm like terrible right because I like to overachieve and I want to be much further along and be kind of a drowning man instead. But if you look back at what you've done you've done some pretty amazing stuff. 

Brian Gregory: [00:41:52] So you can't look at it like that you know you you may only be 20 percent of the way there to your mind but. You know it's a great deal for what. You know when when you really look at the scope of things and I think that. 

Brian Gregory: [00:42:04] I find myself even to read it reiterate that same quote to people on my team right there when they get their heads down and they're like oh why is everything so hard. 

Brian Gregory: [00:42:13] Because it is hard. It is not like that doesn't mean it's not going to be more rewarding in the end but it's very difficult. So I think having you two to talk to people who have been through it and are going to understand it very you know it helps you a great deal. 

CotŽ: [00:42:27] Yeah well that's a good point to wrap up on a high note and I think I think you're right that's what I when I talk with people about this. One of my personal goals which I don't really tell them until I sniff them out is like the stuff is hard. 

CotŽ: [00:42:42] So get ready for disappointment. 

Matt Curry: [00:42:45] Right. 

CotŽ: [00:42:45] Like like as long as you know it's going to be it's it's kind of analogically like people try to tell you how hard being a parent is. 

CotŽ: [00:42:56] And they they don't tell you enough like it's going to be very difficult. 

CotŽ: [00:43:02] And so you need to understand that it's not it's not all just like a bunch of special family ties episodes that resolve themselves in 42 minutes or whatever. It's much more difficult. But hey I appreciate you being on with us. Well we'll have to if you'll deign to do it we'll have to have you on for a second part because I think between Matt and I and even even some questions I've written down while we've been here there's some topics that we should get onto or we should talk about you as well. But you know I think I think that was that was very helpful for the people that we had there. 

CotŽ: [00:43:33] Do you have any you have any like Twitter accounts or your elves you want to hand out to people if they want to follow up with you. 

Brian Gregory: [00:43:41] All yeah I'll shoot those to you. I don't have them. 

Brian Gregory: [00:43:44] Sure. I'll put those changed my name. OK. Yeah. Yeah that we should definitely follow up. I'll be glad to. I would actually love to. One of the exciting things I'm talking about now is the whole funding model and how is it different. How do you treat it going. You know in space Firstly the legacy mindset right because it's completely different. So that's something you like now. So that's kind of exciting. 

Brian Gregory: [00:44:06] Yeah. No I think I think if there's two areas they're not the only ones that distinguish enterprise problems so to speak in the cloud space but that's one of them is is how do you deal with the CFO and funding and thinking about all of that. And then the other one which I'm sure you would have. I mean actually you know both of you guys. Any enterprise has this is basically how do you deal with like regulations and auditors and all of that. And I think it seems like there's almost if I have my shape's right there's a convex answer to it like there's a whole there's a large set of problems that are easy which are sort of like like when you look at when I look at like PCI standards it's kind of like don't do dumb stuff like writing your password on a post-it note right. 

Brian Gregory: [00:44:52] Right. 

CotŽ: [00:44:53] And but but then there's some really hard things that can kind of laugh foul stuff up and there's not really any middle problems there's either really easy problems or really hard problems when it comes to audit and compliance. 

CotŽ: [00:45:05] And I don't I don't think a lot of the consumer space people have those issues so much as opposed to enterprise areas. But yeah we have plenty of topics will we'll we'll have a follow up it sort of. I wouldn't call it a grilling. Just a slight smore cooking or delightful. 

Brian Gregory: [00:45:22] That sounds good. 

CotŽ: [00:45:23] All right. Well as always this has been the lords of computing podcast. You can go find this episode and the other ones that that Matt and I have done in the future ones that we'll do at Lords of computing that come up or if you just go to Kotaro Io which will forward you to you can find this podcast and we'll put some links to Brian stuff in the show. It's as you said and some other like a summary of what's been going on here and in this episode and links to other interesting things including that that 20 percent discount to the C-f summit. If you want to I'll be there and you can come see Matt talk and other people discussing things like this I think they do him in like 30 minute chunks. Right Matt. Or is it 45 minute sessions that they do. 

Matt Curry: [00:46:04] I think the session my session is 30 minutes. I think it depends on the session type. I think that he notes maybe a little bit longer and I think they have these birds of other sessions were really cool. Were people related to similar topics get together. Not sure the time from those ones but the schedules out there were out there. Take a look. I think there a lot of cool topics. It should be fun and exciting summit and then did you want to plug the cotŽ did you want to plug the spring platform conference as well coming up. 

CotŽ: [00:46:35] Oh sure. Yeah. Yeah we have that and I think it's like August 1st or 4th or something like that. I'll put a link to it but it's the let me see if I get the name right. Spring one platform not platform spring one but it's basically pivotal is conference about obviously pivotal Cloud Foundry in spring. But I think I think a fair a large amount of like technical stuff like you know developer things like that but there's also a lot of culture and business stuff. The kind of stuff we talk about here that will be going on there and that they actually have just thinking they haven't been speaking what they suckered me into. 

[00:47:11] That's. 

CotŽ: [00:47:12] Wonderful. Well your are your friends and people are doing their job well. Give it to me myself if all three of us will be there and I'm sure they'll be you know if if you've seen lots of the videos from the Cloud Foundry foundation you know people like like Matt Metts buddy at Allstate Andy have done and other people talking about why they're doing this and the challenges that they have we should have a fair amount of that as well. It's spring one platform. If you've listened this far to the podcast you'll be interested in both those conferences. And with that we'll wrap up. Thanks as always for listening and we'll see everyone next time.