 And we have another new user who's going to be speaking for the first time in the finance industry. This is TD Bank. And what I think is really interesting about the story that TD Bank has to tell is it's not just about technology, but it's about what they need to do inside of their company and what they have done to address the cultural challenges that technology shifts bring. So, help me welcome from TD Bank, Graham Peacock, VP of Engineering. Morning, everybody. You know, I was standing offstage in the wings there, waiting, thinking, why did I volunteer for this? And then it occurred to me that if only Mark and I co-ordinated our wardrobes a little bit better and I'd worn blue jeans, then he could have stayed on stage and done my segment as well. Maybe next year. Okay, so let me tell you a bit about my history, because it's a little unusual. I spent 25 years of my career building software in very aggressive, very high-moving, very complex environments. So my past couple of positions, I've been CIO for credit and emerging market businesses at UBS, at Citigroup, and most recently at Deutsche Bank. So you're kind of asking yourself, you know, why a software development guy is now in a major infrastructure role at TD Bank. And I think the answer is very straightforward. Infrastructure is becoming software. And who better to manage through that kind of transformation than someone who's built software all of his life? So that's me. Let's move on quickly to take a little look at TD Bank, because this is a really impressive story. Top left-hand corner, you can see the geography of TD Bank around about 2002. So dominant banking in Canada, number one by market capitalization, you know, $19 billion, very, very successful company. And what happened then was, you know, the strategy changed to look at expansion into North America. And that really takes us to the picture at the bottom left. So you can see over that, you know, 10-year period how we've penetrated into the U.S., largely through a strategy of acquisition or divestiture, you know, we very smartly chose to get out of high-risk proprietary trading credit products just before the market dislocation in 2008. And as I said, I was at Citigroup in 2008. I was CIO for that business. The day it lost $46 billion, right, very, very smart move by TD Bank. So today we're a lower-risk retail-focused bank with a franchise dealer, and that dealer is there to effectively execute volume from the retail and wealth businesses. And again, those green dots at the bottom left-hand corner show the location of every branch that we have across North America, about 2,500 now, and our market capitalization has grown to $93 billion. We're now ranked number six in North America, and that means we're competing not just with Canadian banks, but with Chase, with Bank of America, and with Citigroup. It's a very, very impressive story. But moving on, we can see that the kind of dramatic growth that we've had and the culture of growing through acquisition or divestiture has actually driven a project-based culture in the bank, which is perfectly natural, but when you're driving very, very aggressive projects against tight deadlines and typically with an acquisition, you know, the dates for the project can't move under any circumstances, then you're making transactional decisions as part of each of those projects. That means you're making design decisions, you're choosing infrastructure, you're making all of those decisions very transactionally on a per-project basis. So absolutely the right way to grow the business. I'm not sure we could have gotten to this point in the bank's history without that kind of behavior, but I think there are consequences. You know, as a result, TD are pretty vendor-centric. The succession of projects that we've run over the years have delivered a lot of platform diversity into all of our data centers. I think there's a lot of vendors in the audience. I think it's almost certain that we bought every one of every product that you've ever made. And you know, that kind of diversity means that there's a lot of one-offs, there's a lot of customization, and there's very little automation. You know, you're driving aggressively towards a project date, and, you know, clearly it's more compelling, it's much easier, much cheaper to throw bodies at a problem rather than invest in automation when you're not thinking beyond, you know, the acquisition date. So that's led to the landscape that we have, which is highly custom, and is relatively little reuse from project to project. So the objective clearly is to address the diversity that we have, to reduce the number of products available to the development teams across the enterprise, to move to a very much more standard, commodity, cheaper environment with dramatically more automation and a self-serve environment. So, you know, the drivers clearly have increased agility and dramatically lower cost, so faster, better, cheaper. I think is, you know, the reason the title on this slide is Moonshot. It is an awful lot to achieve, and, you know, over the next five years we've got an awful lot to deliver. So how do we gear up for that kind of strategy, that kind of initiative? And, you know, normally there's a series of things you would do in the preceding years to address the sort of things that I've spoken about, right? You would try and standardize on the platform catalog. You would try to create standard reusable design patterns that you can then move on to automate. You would use a new technology introduction process to better select the sort of technologies, the sort of capabilities you want to introduce across the enterprise. More importantly, you contain the use of existing technologies and start eliminating some of those that are more proprietary, more custom, more expensive, and focus the entire organization on fewer, more open, more standard, more commodity technologies. Coherency is an important question, right? Because whilst you're executing projects in a number of vertical business lines, you're less focused on standardization. You're less focused on not having 27 different versions of Tomcat, which we do, by the way. But you also are working with vendors to actually have vendors adopt bank standards and bank versions of their standard products, and that's something we've just begun to do. Now, typically, as I said, you do all of these things leading up to the sort of initiative that we have today to deploy cloud across an enterprise of 80,000 people and 10,000 developers. So there's an awful lot of organizational and cultural change required in the preceding years that we didn't have the luxury of working through. So now we're at the point where we're actually we're running this particular process in parallel to building out cloud and forcing cloud adoption across that enterprise. Pretty daunting set of activities. Moving on, in terms of how we manage that kind of cultural change, the approach we've taken is really one of, you know, how would the software industry choose to organize itself or how does it organize itself today? And why can't we take that same philosophy into a bank and actually deliver software within a bank, an infrastructure within a bank, the same way as the industry does to its clients? So step one is really to get our own house in order, beginning with infrastructure. So clearly if we can control the diversity of the product catalog, the number of versions in existence, get sponsorship from the top of the house through all of the CIOs in the very siloed development community. The second step is really about building the capability within the infrastructure team in order to deliver against that strategy. And here we chose an approach that's a little bit against the direction of the industry, right? For years we've seen developers effectively move to medium or lower cost locations. Now, in order to build an engineering team within TD against the timeframe that we're working towards, we did the reverse, right? We consolidated our locations, we moved away from lower cost locations. We actually created an engineering center in Manhattan. In the first three months of this year, we hired 30 people into that office. By July, we'll have 50, right? And we're doing that for a very specific reason. We need people with truly international experience that will solve some of these same problems before more than once, right? Trial and error and repeat. Won't really get us where we need to be as quickly as we need to get there. So clearly that engineering team that is split between Toronto and New York is there to execute that strategy, execute that roadmap, and remain very, very focused on the product development side of the cloud program. So that's the dot-com within the bank. The second step is to remember that, you know, there are 10,000 people in technology at TD and we're dramatically outnumbered in terms of the ratio of developers to the ratio of engineers in the infrastructure team. So we're approaching this the same way as it vended us. We need to sell the benefits of cloud constantly. We need to train a very, very federated technology team across multiple cities, multiple countries, and multiple technologies. We need to coerce them into moving to a dramatically shorter list of technologies and versions to move to cloud. But training is an interesting one. You know, we stood up cloud very, very quickly in mid-December. In January, we started a training program. We trained 600 people in January. Year to date, we've trained over 1,500. We're now running bi-weekly open mic sessions. We're publishing bi-weekly FAQs. Engaging with the community is a very big activity and one that needs to be continuous. Remember, this isn't a short-term project. To migrate all of our applications to cloud will take five years plus. You know, the target is to get 80% there in the next five years. So enabling and empowering and supporting the development team is key. So how do we sustain it? So the cloud first strategy that we have is a great one. You know, every development team in the bank needs to put their project live, their application live on cloud, unless there's a very, very good reason that they can't. And if they don't think they can, they need my approval to build a traditional physical server or a virtual server. And that's pretty hard to gain. Now, if people are running projects in the current quarter, the current calendar year, it's relatively easy to steer them to the cloud. But to be honest, you know, my team doesn't really think beyond that. I can't predict whether businesses are going to invest their money next quarter, next year, or in five years' time. So we've spun up a separate cloud adoption program that is actually actively working across all the business lines in order to help them understand their application portfolio. And there's about 4,000 applications in the bank, so it's a pretty diverse portfolio itself. But that means I tease up the pipeline for future years, future quarters. That means the business lines have to work with us to decompose their application architecture, to understand the sequence that will move those components to cloud quarter by quarter, year on year. And that means we can hold them accountable for delivering against those commitments, which is pretty key. So that's a quick walkthrough of the journey at TD, some of the techniques that we're employing to deploy cloud to the bank. Much more comprehensive session we have scheduled on Thursday. And some of the technology choices we've made along the way are pretty interesting. Some of the partnerships we've created and forced across the open source community are actually proving invaluable at this point in time. So please join me on Thursday and we can walk through the technical detail behind this program in much more detail. Okay, thank you. Thank you, Graham.