We have Salesforce, and we have Alice here, who's the EVP and General Manager of Platform Integration and Automation. Alice, thanks so much for being here. Thanks for having me. Awesome. Alice, maybe to kick it off, maybe you can just start by giving a brief background about yourself and then your role at Salesforce, too. Yeah, so I'm the GM for the Platform Integration Automation at Salesforce. Before that, I was at Microsoft for a long time, worked on a ton of different teams there, was at Code.org, working on computer science education. I'm really passionate about developers and creators and enabling people to build. Working on a platform, for me, I love it, and I love working on the Salesforce platform. I think right now we're at this moment in time where AI is dramatically changing both what we can build on the platforms and how we build. It's a really fun time to be at Salesforce and working on the Salesforce platform. Perfect. Awesome. I'm sure not everyone in the audience is familiar with what's exactly included in platform. Maybe you could just take some time and break that down for us. Yeah, so it's confusing. I'll give it to you. Basically, there are multiple pieces of it. Part of it is the shared application platform that goes inside all of our clouds at Salesforce. Things like authentication or the mobile platform that's part of sales or service or industry clouds, marketing. There's another piece of it, which are the horizontal businesses that we run independent of what cloud you buy. My team supports Privacy Center, Security Center, Shield, all of these horizontal products that apply to every single one of our clouds. One of the things that I'm excited about is my team also supports the developer platform. That's everything from getting Salesforce and customizing it for your business to our SIS and being able to support all of the different customizations that they do on the platform. Also, companies build their own full applications on the platform. All of the application development that runs people running in their finance teams and their back office and operations, all of those apps, there are 9.5 million apps built on the Salesforce platform. All of those are supported by my team in the developer platform. We also have a huge ISV network on AppExchange. All of those companies are also built on that core Salesforce platform. My team also runs integration and automation. How many of you know MuleSoft? Yes, lots of hands here. Being able to connect all of the different applications in your enterprise, which I think is even more relevant right now in this AI era, because now we're using that same capability to talk about how do we connect agents across my enterprise? How do we allow agents to talk to each other across all the systems in our enterprise? The integration space is now the agentic integration space. Right, yeah. Let's take that a step further, and let's talk about how that relates to Agentforce. Highly competitive space that Agentforce is in right now. What differentiates Agentforce? Maybe you can talk a little bit more about the deeply unified platform and how that gives you guys a competitive edge and also how you're leading with trust in your efforts there as well, too. Oh, that's a lot. Okay, I'm going to break this down. Lots of questions in all in one. That was like four questions at the time. I am going to break it down. I think we talked a lot about it on our recent earnings call. We talked about this A.D.A.M. framework of the thing that differentiates us is that you bring together the apps, the data, the agents, and the metadata platform. That is pretty abstract. I am going to try to make that a little bit more real. I think we all use LLMs and AI all the time. I actually used it to plan a dinner party this weekend. Oh, no way. I did. I did. It told me all the ingredients I needed to buy and pulled together a bunch of recipes for me. It was pretty great. All of these consumer AI products are basically feeding back to us a bunch of public data. And they're trained on basically the same set of public data. They're all trained on the internet. That works great if the question that you want to ask is a public data question. This is all the consumer AI. That's amazing, right? If I want to know how to feed eight vegetarians something good for dinner, that's public information. I can just go grab that. I think the challenge is when you go to look at how do I roll this out in an enterprise to solve enterprise scenarios, it's all about my data, right? Finnair is one of our customers. When they're looking at the airline data, yes, what flights are up are on the public internet, but not what flight I'm on, not my information. There's a lot of PII there. There's a lot of internal data that these companies are working with. When you take the LLM and you say, how do I apply this? How do I make it agentic within my enterprise? There are multiple problems you need to solve. First, you need to solve the data problem. You need to understand your enterprise data. That's the structured data. It's the unstructured data. It's the metadata, which is the data about the data that understands what that data is. Even if you solve all of that, you get all the data together, you expose it all to the LLM, you do all of the RAG stuff that you need to do, all you've got is a chatbot that can answer questions. It's a Q&A bot. If you really want to get agentic, the next thing you need to do is you need to enable that agent to be able to take an action in your organization. You need to be connected into the actions that you could take. We have built in the capabilities to take action across sales, across service, across marketing, across commerce, across all these industries. They're built in as part of the Salesforce platform, that deeply unified platform. We also have this rich automation framework, which allows companies to build their own specific automations that connect in. With MuleSoft, we connect out to the rest of your enterprise, to Workday, to SAP, to all of these other pieces of your enterprise so that your agent now has the right enterprise data and can take the right actions across your enterprise. That is all connected into the LLM. It is all connected into the trust layer that you need in order to run this in an enterprise. Enterprises are also dealing with compliance. They are dealing with auditing. They are dealing with the old school stuff that we have been dealing with for years, things like, do I have the right security permissions? Not just you, not just Taylor. Does my agent have the right security permissions? We need to take all of these rich capabilities that have allowed Salesforce to run in regulated industries and banks and all of these places, but now apply them in the agentic era. We need to add new capabilities like guardrails and red teaming around the agentic capabilities. Our differentiator is that we've built all of this together in one platform, one open platform that's connected to the rest of your enterprise, which accelerates the time to value for our customers. It enables them to take these amazing capabilities, these amazing LLM capabilities, and put them to use to solve real problems in their enterprise and to do it fast. We have 800 customers that are already live with Agentforce. For anybody here who's been in enterprise software for a while, that's amazingly fast, right? Enterprise software is not something that usually goes out that quickly. It's only been live for two quarters. I think it's that core capability, it's that deeply unified platform that enables our customers to find value quickly. Yeah, awesome. Now that you've had Agentforce live for six months out, maybe you can talk about some of the use cases that you've seen that are gaining the most traction early on. Boy, there's so many. What are some interesting use cases? I think one of the things that we've done is we've made it easy to get started with some of our core use cases, like sales or service. As a starting point, I think of it as a maturity model for companies that are embracing these AI technologies, where the easiest place to start is to say, okay, I'm using some of those inbox functionality. I'm going to use it to answer support questions or answer sales questions. Even with just that, you can get a lot of value. You can get a lot of value, and you can get it quickly. I'll give you an example, 1-800Accountant. I can't even imagine trying to run a tax company during tax season. You have to onboard so many people. You get this flood of all of your requests come at exactly the same time. You're going to get this flood of requests coming in. And they're not easy answers. You have to train up a staff of people who can answer whether or not I should get a deduction for my car because of this or that. That's challenging. It's a challenging space to be in. What they did was they used Agentforce to be able to handle that influx of caseloads. 70% of their cases in this last tax season were handled autonomously by Agentforce, which means you don't have to onboard and train up as many people, but it also means that their staff can spend more time on the high-value conversations. I think that's a place where a lot of companies start. They start with that base use case of, I'm just trying to get my existing system to work better, work faster. Once they do that, what you find is you open up a whole bunch more scenarios. You start out with something like support. One of the cool things about agents is we talked about training staff, right? They're not siloed. They can understand things outside of the space that they're in. Instead of having a support staff over here and a sales staff over here, where when this case goes from upsell or I want to do cross-sell, maybe I move it over to another person who knows how to do that. I can instead give the capabilities of doing that cross-sell and upsell to the same agent that's supporting me in support. Now I can get more value from that agent and start bringing in new scenarios to the agent. That is where I see a lot of companies going next, adding additional capabilities into their existing agent. Those capabilities could be more complicated support things like, I want to do order management. They could also be new capabilities that are outside of where they originally went. The other thing that we see companies doing is extending out to different use cases and different personas. So reMarkable. Have you used a reMarkable tablet? No, I don't think so. Oh, they're fantastic. Okay, for those of you who don't have one yet, these tablets are great. The best thing about them is they're not fully connected to the internet. I mean, they're connected to the internet, but they don't have all of those distracting things. You can actually, they force you to focus and think, which is, I think, beautiful. I'm a huge fan. Anyway, so reMarkable tablets rolled out. This is, again, that speed of ROI. They were able to roll out their first agent externally facing in just three weeks and already start handling customer cases. They said, hey, this is really working for external customers. Can we use this for our employees? They built an employee agent, which they called Mark, which is cute, reMarkable. It's Mark, the employee agent. Their employee agent is on Slack and answers internal questions for their team on Slack. I see people doing that where they start in one scenario, and they're like, oh, we can use it over here. We can use it over here. We do it internally. We've been rolling it out across the board, across all of our own use cases, because everybody at Salesforce is like, hey, I want to use it for this. I want to use it for that. Right now, we're doing our promotion process. I was just yesterday submitting my promotions for my team. We used to submit them through this form online. Now we do it through an agent. I went to my little agent in Slack, and I talked about the promotion and it asked me questions. I was like, oh, but when do I need to know this by? I had a question that was sort of outside of it. I was just able to ask. The agent gave me an answer and helped me submit my promotion request. That's awesome. Yeah. That's great. Let's talk about where those use cases are in terms of getting to the production phase. Kind of the last time that we did a bunch of Agentforce checks, it still sounded like there was a lot of customers that were still in that discovery mode, right? When you speak with customers, where are they in terms of getting to that production at scale? When can we start to see most Agentforce customers on the platform really operating at a bigger scale than maybe the discovery and some of what we're seeing today? I mean, I think you're going to see both. It's enterprise software. There's a lot of companies who, for a lot of good reasons, want to do a proof of concept. They're going to want to test it. They're going to want to try it out in a scoped area and roll it out in a deliberate fashion. We're going to see, and then we're going to see companies that move a lot faster, like reMarkable that I just spoke to. What we're seeing is, I think, frankly, incredible, the traction that we've already seen with this. 800 customers in production is incredible. A lot of these customers are at massive scale already, which is just incredible. I guess Smartsheet's a good example of this. Smartsheet is a how many of you use Smartsheet? I've got a lot of people here, but not everybody. Smartsheet is a fantastic tool for managing projects, estimating budgets. They have about 13 million users. They had a help agent outside of just like a typical, okay, you can go get help. What they saw when they started to pull out Agentforce was not only could it answer questions and help deflect cases in a typical support fashion, but when we talk about scale, they could expand the scale of what they were doing. They do not need people to go to the help website to go get help. Because they have autonomous help now, they could offer people a lot more help. They actually took Agentforce and embedded it inside their product. Inside of Smartsheet now, I can say, hey, I need some help on what does this formula mean? Agentforce is right there inside of the product, available for all of their users. It also means that, again, this is that crossing borders, right? In the past, I would have said the only thing I could have done inside the product probably was maybe get help about what the formula did. Now you can get better help about that. Also, I could not have gotten sort of customer support help, right? I can do that too. I can ask you questions about my user permissions and provisioning and those pieces. You are bringing together these different silos of information, and you are putting them together at scale inside of the Smartsheet product that is already shipping to all of their customers. Yeah, that's really I used to cover Smartsheet. I didn't know that they were using Salesforce for some of that behind the scenes. That's interesting. In terms of big, big news, obviously, the Informatica acquisition is the biggest news this quarter. Maybe you could just take a moment to describe how that fits into Salesforce's AI vision and what capabilities Informatica brings to Salesforce and how it's going to enhance the platform overall. Yeah, we're really excited about Informatica and bringing that into this deeply unified platform. I think it goes back to what I was talking about at the beginning, data, right? This is the foundation of this agentic future is understanding your data. Everything you do with AI depends on the data that you're bringing into it. Informatica is world-leading at understanding and bringing together the data within the enterprise. They bring in rich capabilities in data governance, in data quality, in MDM, in metadata across my organization. I mean, it's not just what is the data, but they even have rich data lineage, right? So I can understand the history of this data and where did it come from. Bringing that together as a core foundational piece and connecting it into the platform, exposing that rich understanding of the data into Agentforce, I think, will supercharge all of our capabilities. It also complements our MuleSoft portfolio, bringing together the data integration and then being able to expose that rich understanding and governance of the data as part of the application integration that we do in MuleSoft. Of course, Tableau is even more exciting when you can do analytics on that rich set of data that Informatica makes available. We are looking forward to that closing. Perfect. Awesome. Maybe you can talk now about some of the synergies that you're aiming to drive through the acquisition and some also too that you've established with the partnership already. Maybe you could just elaborate a little bit there and how your role plays into that as well too. Yeah, so my role personally, I'm most involved with it from the perspective of MuleSoft working together with Informatica. We're really excited about that and how that's going to work together. I'll talk about all of our all Salesforce because I think there's a lot of opportunity with Informatica to accelerate a number of different areas across Salesforce. We talked about data governance and data understanding. When you take that capability and this understanding of metadata that goes beyond the data that we have in Salesforce today, Data Cloud today is a data lakehouse on Salesforce that enables you to connect to all of your data across your enterprise and expose it to take action on it in Salesforce and all of our applications and in Agentforce. That capability and what it does also is it allows you to unify and harmonize that data around the customer persona. So really understanding the 360-degree view of who is my customer and then being able to take action on that across the Salesforce properties. Informatica brings a richer understanding of all of the data across my enterprise and other concepts with MDM across my enterprise beyond the customer. So I can understand my suppliers, my business at different levels across the entire enterprise. I have deeper control and governance across that metadata in my enterprise. When I talk about the data, I'm simplifying it. When I say, hey, we want to bring the data together for Agentforce, that's just one piece of it. The data is part of it. You also really need another for enterprises, you need another level of understanding of that data, the metadata around it. What is that data? What does it mean? How is it being used? What objects within the system is that data a part of? What influences what? Where does it come from? What Informatica does is it brings that rich understanding, which we're then going to use to enhance Data Cloud and also to enhance Informatica. I think together, it's going to be even more powerful. It also complements our MuleSoft portfolio, where we provide, we enable today all of these applications to connect to each other. Now with AI, we're allowing agents to connect to each other with MuleSoft. We support A2A and MCP so that you can now have an agentic enterprise connected through MuleSoft. All of that, again, is based on the data. A deeper understanding of what that data is, being able to connect to that data, makes all of those integrations more powerful and allows you to do those integrations in an even better, more governed way where we can bring in the data governance into the MuleSoft portfolio. We are looking forward to that. Tableau is now going to be able to take this richer insight into my data and pull that forward when I do analytics on my data. Yeah, perfect. Salesforce's portfolio has expanded materially over the last several years with Tableau, MuleSoft, more recently Own Backup, not to mention the additions to Data Cloud and Agentforce as well too. Can you talk a little bit about the product strategy and how all these pieces fit together? Oh, that's another big question. Okay, there were at least five companies. Which ones were we talking about? I think some of them were before this recent agentic revolution, and some of those have been after. More recently, you see our investments in the importance of data in the enterprise in Informatica and in Own, which offers data resilience and helps companies manage their data, which I think, again, is really critical and important in this agentic space. I'm also really glad that before LLMs hit the scene, we had made these investments and some really strategic investments in MuleSoft, in Tableau, and in Slack. I think those are three recent investments. I guess just breaking it down, let's talk about MuleSoft, Tableau, and Slack. How do they fit into this overall strategy? With MuleSoft, I talked about this. What does it take to build AI in the enterprise, right? There is the data, and then there is this action layer, right? How do I take action across my enterprise? I think for years, MuleSoft was solving a problem that's true to every enterprise where I've got 900 different systems, and I need to figure out how they're going to talk to each other. I want to create reusability. I want to have a governed system so I have control and management and audibility of what's happening across my system. MuleSoft provided API management. It provided the governance, and it created that connectivity and reusability across my enterprise. Now we're moving to a new world where all of those same problems exist, and they're supercharged with AI. Because I now just have deterministic systems connecting to each other. I now have agentic systems connecting to each other. MuleSoft has been moving, I think, is becoming a more relevant play because it's enabling us to connect those agentic systems with all of those same capabilities that we needed before around reusability, understanding, governance. You just need those even more when you have a bunch of AI. We've already launched support for MCP servers as part of MuleSoft. All of those APIs that you built in MuleSoft are now you can just expose them as MCP servers. You're already agentically ready for this next generation with MuleSoft. Now you've got all of these capabilities that you've built, these tools. In the same way, I would have said, hey, my APIs are the language of my organization. These are my tools or my capabilities of my organization. Now I can expose all of those and say, those are my agentic capabilities of my organization, right? And then we've updated our Flex Gateway to be able to support governance as I'm calling all of these AIs, right? Because I don't just want anybody to be able to call an agentic capability, either whether it's A2A, the agent-to-agent protocol, or MCP. I don't want you just to be able to call it unlimited, right? They're expensive. It can run up a bunch of bills across something, right? And I'm not saying anyone in particular. This is across every single agent I'm running from any different vendor, right? You want to be able to understand the permissions, the controls, the governance, the audibility. All of that capability is built into what MuleSoft is now providing in this next era. We invested in MuleSoft back when it was about connecting all of these applications in my system. Now I think it is about enabling our agents and all of the agents to connect across your enterprise in a way that is governed and controlled. I think that is going to be even more relevant as we move forward. Tableau was the next one you had on your list. I think Tableau is interesting for multiple reasons, both because obviously everybody is bringing this data together. You want to analyze it. You want to understand it. I also think data is more powerful with the analysis of data becomes more powerful because the agents can do it too, right? It can take a long time to sift through data, to understand it, to get the insights from it. The agentic capabilities can supercharge some of this ability to understand. We have a feature called Tableau Pulse. It just slacks me every day with just the one thing I need to know and gives me updates on the key pieces using AI to help me understand my business. I think that kind of capability, we're going to see more and more of that as we move forward. Another thing that we've been able to pull in from Tableau is the semantic model. I do not know if you know this. Inside of Tableau, one of the things that you do, and this is true, this is something that data analysts, business analysts would do, is you define what are the KPIs for your business, right? You look at this and it's a bunch of data. I say at Salesforce, we have numbers we care about, our ACV, our AOV, our net new AOV, right? Every company, even simple things like sales, people define them differently, right? Like, oh, did you actually mean sales in Europe? Did you want to count this department or not that department? You need to put together some logic that you're going to use for all of your internal dashboarding and reports around what are the semantic numbers that matter for your business, what drives your business. That understanding of what are those semantics, we call the semantic model that's embedded in Tableau. Now we've taken that and built it into the deeply unified platform so that that same semantic understanding is now available inside of Data Cloud. It can power all of the work that you're doing, and it can power the agents as well. The agents do not just know the data. They know the semantic meaning of that data. They are able to care about those same numbers that you care about when you are driving your business. There is a third Slack, a Slack, right? Slack is, I do not know, my favorite way to talk to agents, right? It is conversational. It is the way we talk to each other. I live in Slack all day. Now the agents live in Slack with us. We can bring agents into the conversation in a natural way so that it is humans and humans can interact, humans and agents, and agents can join our conversation. We are building all of our core apps into Slack so that I can now use inside of Slack, we can swarm cases, we can see what is going on with sales. We have agent support right there inside of Slack. It is all customizable and extendable. Customers can bring their own agents into Slack. I was talking about reMarkable. Mark lives in Slack. Mark the agent is inside of Slack. Their employees are just inside of Slack able to talk to Mark. I think it is a fantastic place to be the agentic workplace of the future. Perfect. Yeah, that's awesome. Lots of fun things going on. Maybe we'll shift gears a bit. This is our annual Women in Tech Conference. Only appropriate to now ask some women questions. Just as you as a female leader at Salesforce, I guess, how have your experience shaped the way you lead in such a dynamic environment? How do you think about mentoring, empowering the next generation of leaders out there? I mean, anytime I get those, I have to start with a big thank you to the women who have helped me in my career. My strongest support network are women. And they're women who I worked with at Microsoft. They're women who I know. I live in Seattle, not in the Bay Area, but some of them moved down here. So some of my network is now here. They're women who are down here. I think I really appreciate the network of support that we as women have in this space with each other. I want to pay that forward. I want to keep growing that network. I've really liked being at Salesforce. I think Salesforce has a real commitment to supporting women. We have what we call Women in Salesforce. It is a group. It has 20,000 women in it. We do outreach. We've been doing outreach around the world. We just passed in India. We were doing upskilling events for women. We just hit 500,000 upskilling opportunities for women in India. We were at our leadership conference a few weeks ago. I don't have the actual numbers. We probably published them. We do publish them on our website, so you can look them up. There were just so many women in that leadership conference and people who have helped me as I've onboarded, people who have supported me. I appreciate the way Salesforce has really supported its network of women leaders at the company. Yeah, that's really great. Maybe just dive a little bit deeper into that. Let's talk about mentoring, right? Yeah. What's a question you often hear from mentees, particularly in the era of rapid technological change and everything that's going on? What's your advice? okay. Actually, I think roughly a month ago, I was doing a, I also mentor a lot of people, but I was speaking at a women's conference for a bunch of college students who wanted to get into AI. I got this question from them as well. My advice is to get hands-on. I encourage people to actually try out these technologies. I think they're more approachable than we think they are, especially because we have AI to help us use them. If you have questions, the AI will help you figure out how to get through those questions and how to use it. I think for a long time, some of these technology pieces felt too intractable to even get started. These are completely, everybody can use these things. Everybody can do it. You can use it to plan your dinner party. You can use it to plan your trip. You can use all these consumer things. You can also build your own agents. We actually have been running at all of our conferences. We have a set of tables where you can come and build your own agent to solve the challenges that you have in your organization. Now, this is a prototype agent because it's not connected to your data, right? You need to go back and connect it to your data and your APIs, those pieces that we just talked about. We have had over 10,000 agents built at the conference. You can play with it. You can use it. It can answer the questions about the internet, the ones I said were easier, the internet questions. I think the more we can all try it, the faster we can accelerate ourselves to the place where we become the experts. We're all leading this revolution. Perfect. Awesome. That was a great way to end. We will end it there. This also concludes all the panels and keynotes. Thank you to everyone in the audience for joining the third year of the Women in Tech Conference. Awesome. Thank you. Thanks so much, Alice. This was great. Oh, and there's going to be a reception to follow as well, too.