Name: The Future of IT Operations: AI That Knows Your Environment and Fixes Problems
Uploaded: 2026-03-25T20:01:07.454Z
Duration: 41 min 12 s
Description: The Future of IT Operations: AI That Knows Your Environment and Fixes Problems

Transcript for "The Future of IT Operations: AI That Knows Your Environment and Fixes Problems": Alright. Thanks for joining us today, folks. We're gonna give it just one quick minute while we wait for people to keep shuffling in. But definitely feel free to toss in the chat any questions you have. We'd love to hear where you're calling in from or, you know, what what brought you here today. Obviously, there's gonna be some some cool stuff we're talking about, seeing some really brand new stuff in action. It's gonna be a good time. So just a few more seconds and we will get the ball rolling. Alright. We have a good amount of people here. Thank you so much for being with us today. We're talking about something that every IT team IT team deals with constantly, and that's understanding exactly what's happening across their environment and really what the future of IT administration looks like and how it keeps evolving and improving right in front of our eyes, really. So my name is Carter Francis. Joining me today is Zafar Kharani and James Sorrenti, two of my favorite colleagues at Rippling. Zafar is a product lead for Rippling IT, and every time I get to talk to him these days, he's showing off cool features and improvements. And today is one of the more exciting ones in a long line of really cool stuff. And if James is with me on a webinar, we're we're always having a good time. So he's our strategic strategy he's our strategy and community lead. So you might be familiar with him in our Mac admins channel, LinkedIn, or elsewhere. His expertise and insight is always very welcome. So thanks for joining us, gents. I think we've got a lot to talk about, so let's cut right to it. Modern IT environments are totally fragmented. Right? You have identity systems, device management platforms, access controls, security tools, inventory stuff, maybe like homegrown solutions to try and pull stuff together. And they all have separate parts of the story. And that means when something breaks, or when you need to answer a simple operational question that you're pulling reports or hopping between tools and trying to manually connect the dots. So today we really wanna explore the bigger question of what would IT operations look like if AI could actually, like, fully understand your entire environment and help fix problems, not just report them. Right? So software. Let's start with the reality that most IT leaders face today. Why is it so difficult for teams to answer these operational questions? Yeah. You know, Carter, I think you you hit the nail on the head. The core issue is fragmentation. Right? And most IT environments were not designed as one system. They were assembled over time. So, you know, you have an identity provider over here, an MDM solution over there, a separate access management tool, a different security platform, probably a ticketing system bolted on top of all of that. And each of those tools has a piece of the picture, but none of them have the full picture. So when someone asks a seemingly simple question, like, who has access to this app and why, or why is this device out of compliance, the answer doesn't live in any single tool. And so what actually happens is the IT admin has to go pull data from multiple systems, cross reference it manually, and piece together a partial answer, and if that answer needs to turn into an action, then they have to start the process all over again in a different tool, right? And so that's not a tooling gap in the traditional sense, because every individual tool works fine on its own, But the gap is that nobody has the connected view. Right? And without that, IT teams end up spending a huge amount of their time just correlating data instead of actually solving problems. Yeah. Like, IT teams as like point solution experts. Doesn't sound like it's a good time if like only one person knows how to do one thing or doesn't have all the access they need. Right? Exactly. Sometimes we'll hear IT leaders describe things as like swivel chair operations or like like you said, they're just constantly switching between tools, or having to, like, involve random stakeholders outside of IT if they don't have permissions to the right stuff. So, like, James, what do you think from a practitioner side? What what kinds of things and what kind of operational questions that sound simple actually take hours to answer, like, in a in a real world scenario now? No. It's a good question because it really does show up in the, like, questions people ask because I've already kind of seeded some of this about the who was accessed and why did why did it fail and all that stuff. It's like, how do I fix the issue Is the core of it. How do we fix something in IT? But from the employee perspective, so to answer them, you usually are pulling from identity, your MDM, your ticketing system, all those bits, even logs sometimes, you have to go through them, and then you're manually putting it all together, and that's time. It's not that the data doesn't exist, it's just in different places and there's no shared context between them. So the job you spend trying to match everything together is just it's a time suck. Even basic self-service breaks down, we would love when people can help themselves. But an employee can ask, Why can't I have access? Or How do I fix this? The system needs to actually give relevant answers that can help them. So you're not just solving for IT efficiency. You're limiting how many users, like, actually help themselves when you're not giving answers that help them. So most of the time, you're not solving the problem yet. You're just figuring out what actually happened. Yeah. Or figuring out how to use a different tool that you haven't used. Right? Like Oh, sure. Yeah. It's kinda like when you both are, like, talking about someone in IT having to, like, manually put data together, basically. Like, what actually happens behind the scenes for for that to happen? James? I mean Yeah. I mean, yeah, you start building workflows, like, in like, workflows in your head, how to make these things work, and they're in your head. Right? So you check one system for identity, and then you another one for the device, and then another for access. And you know that every time this problem pops up, you're gonna do this, this, this. And it's, you know, it's easy steps, but there are multiple steps in multiple places you need to get in order. Right? And then not everybody on your team might know those like, it's in your head, like you said. Right? Oh, no. And I I look both ways before I cross the street to make sure that I keep that context going. Yeah. But, you know, you have all those data. You're lining up time stamps. You're figuring out what people did what and where, and it's I don't know. You're kind of like a a sleuth or an investigator for, like, regular IT problems when a user is just trying to get their job done. So, yeah, your answer isn't just in one place. It's just all over the place, and you your brain, your process is all over the place trying to figure it out. So you're translating everything, and then you have to then also, after you translate the systems into your context, have to translate for the user too. That's a whole lot of context switching. And, you know, and you're just still making a best guess to the end. So, yeah, it might just sound like a quick solve, but those minutes turn to hours real quick. Yeah. Manual stuff gets deep real quick. Well, obviously, AI is popping up everywhere in IT right now. But most of what we're kinda seeing in the IT world is is stuff that, you know, I I feel like a lot of folks in our industry have have been a little leery of, you know, stuff like help desk bots or, like, ticket summaries or report generation, like, stuff that's, you know, sometimes very useful, but it doesn't really solve, like, operational challenges. Right? So, Zafar, can you brief folks on, like, why has AI struggled before now to actually help with, like, IT operations? It I mean, in short, it's because most AI in IT today is simply working with the wrong layer of the problem. Right? And if you look at what's available right now, like you said, chatbots to deflect help desk tickets, tools that summarize logs, and maybe even systems that generate reports just a little bit faster. Yeah. And and like you said, it's can be helpful. Right? It reduces some friction. But it doesn't solve the operational challenge, which is understanding what's happening across your entire environment and then doing something about it. Right? And and the reason it falls short comes down to, in my head, at least three main things. First is what we've already mentioned, but it bears repeating, right? And that's context. Most AI systems don't actually know the environment in full. They don't know who your employees are, what devices they have, what policies apply to them, what access they've been granted, and so without that, the AI is just guessing or pattern matching on, like, generic data that's not for your specific environment. So that's context. The second reason it falls short is just trust. Right? Even when AI produces an answer, it's really hard for IT teams to verify it. Right? So, like, if the AI says this user has admin access because of this policy, but you can't trace that back to a real policy in the real system, then it's not particularly actionable and you still have to go and check annually, which defeats the purpose. Totally. And then the third reason in my head is that it falls short just because it's really hard to take action today. Like, if the AI diagnosed a problem, and let's just assume for a second that it diagnosed it perfectly, most AI today can't safely do anything about it. Right? It can't actually then, like, you know, maybe you need to revoke someone's access, or you need to somehow remediate a compliance gap or or stage a fix for review. And so maybe the AI will tell you what the problem is, and maybe it's correct. But then you again have to go back to doing this all manually to actually implement the fix. Yeah. Yeah. It's it's like if you don't have three, why would you even begin to entertain I mean, you don't have one and two, why would you even begin to entertain three? Right? Like, if you can't Right. Have that context and trust that, like, the last thing you're gonna do is, like, volunteer to plug something random into your MDM, I imagine. Yeah. I mean, I wouldn't do it. A 100% folks on on on the call today wouldn't do it. Yeah. So, I mean, the issue isn't just, intelligence or, like, a technical limitation or technical possibilities that can't be reached yet. It's that it's really that AI doesn't have the right system context to work with. Yeah. Yeah. That that's exactly it. Like, most AI and IT today is, like, sitting on top of one system at a time. So you're still connecting everything. You're still putting the lines as being the pins on the map you built for your problems. It could summarize your tickets, generates a report, but it doesn't actually understand how it all, you know, relates. So that that context you've been mentioning, it's can't really answer the why questions, and it's definitely can't take action a way you can trust. You know? Yeah. That's that's pretty much it. Yeah. So I guess then that brings us to kind of our big reveal. Like, let's let's explore and talk about what a different model looks like. Like, what if AI could actually understand the relationships between your employees, their devices, all their access and permissions across the whole environment. Right? Zafar, what what does a unified system of record do to change what AI can do for IT teams? Yeah, you know, I mean, it's really exciting, because it truly is a game changer. And it's a game changer because it addresses the three problems that I I just mentioned. Right? Like, the the piece on context, which then leads to better trust and the ability to take actions. And it and it addresses all of those three things all at once. Right? So when you have identity and devices and access and policies and employee data in a single system, then the AI doesn't need to guess or stitch data together. It doesn't need you to kind of stitch things together manually. It already has the full picture. Right? It knows who the employee is. It knows what team they're on. It knows what devices are assigned to them, what applications they can access. It also knows all the policies that govern all of that. So the AI can answer complex operational questions, and it's not just by summarizing a report, but by actually tracing through the real relationships in the system because it is one system. Right? And and because it's actually pulling from the actual system of record, the answers are verifiable. So from that trust perspective, you're not looking at a hallucination or best effort synthesis from a bunch of different data sources stitched together. You're looking at the real state of your environment that you can actually verify right then and there. And that turns AI from this kind of, you know, search engine, help me like do little bits of my job faster into something that's more closer to an operational analyst that genuinely knows your environment and can show its work so that you can trust. So stuff like, I'm not going to go in and manually export a report, I'm going to ask Ripley AI for a report with exactly what I need formatted how I want. Or if I'm gonna say why I would actually go a step further, instead of just exporting a report, an IT leader could basically say, well, like, what's the reason why they're exporting a report? Is it because they want to understand why a particular policy like maybe didn't work or didn't apply in the way they intended, they could just ask that second order question. Right? Like, why why yeah. Why even do the first part if you're looking for one answer? Right? Yeah, exactly. Yeah. Like, why did this policy not do what I intended instead of, hey. Can you export this report to me so that I can manually figure out the answer to that question? Right? That's that's the that's the level up that I think we're talking about. Yeah. That's a that's a big shift. I find myself, you know, depending on what AI I'm using and why I'm using it, very much, like, catering my prompts based on, like you're saying, what I understand its truth and scope of context to be. So, like, knowing that it could touch everything in my rippling environment is like a a a pretty big paradigm shift. James, from, like, an IT leader's perspective, folks you've been talking to, like, what really would that would that change? How big of a shift that's is that gonna be for them? I mean, it's pretty big for sure. I mean, it's it's different. Like, you're not asking AI to summarize segments anymore. You're asking it to reason over a system that reflects actual reality. When those pieces are connected, the question becomes something the system can answer reliably and do more than answer. Right? Like, you were saying it wasn't just the reports. It was the, like, remediation. So it's not just what happened and why it happened and what should we do. From an IT leader, that means a few practical changes at hand, like less time chasing contacts across tools, obviously. Right? But also faster reviewable faster reviewable remediation. Like, instead of long investigations, you don't have to be that sleuth anymore. And more confidence in the answers because you're getting data right out of the system with the directly tied to the data points. So it ends up being less about getting info out and more about the system actually helping you, like, operate and be that leader you are. Yeah. Like, asking even something like, tell me if any of my users have laptops that aren't encrypted. Sure, that's an easy enough thing to currently do in rippling. But, like, if you are not if you're someone who's supposed to be able to see that information, but you're not the day to day IT guy, then, like, it might be an entry. Right? That's that's an interesting thing to propose. It's more than just show me it's more it's more of the do it. Right? Like something about it. Yeah, exactly. Sure this team's computers are all encrypted. And probably there's remediation step, we have to do a thing to make it happen, because there's a human in the loop. But like, it'll do the process. It's not just telling you stuff, which is kind of a step forward. Yeah. Moving from just answering questions to like, help fix our these, this is gonna help fix problems. So yeah, I think from my perspective, the mindset shift is, this is not a tool to help you do your work. It's a tool that does the work. Yeah. Right. So you can focus on the higher level strategic things. Interesting. Curious how you train the people to ask it to do that work instead of just ask the questions because we're just used to it helping us Yeah. Just go limp limp along with our our keyboards. But it's just do it has access to the actual nose and noodles putting things together. That's kind of that's that's a shift. That's the shot. Do you have, like, some example, like, operational workflows offer them? Like, what, like, what, what do you see this helping you with the most? I mean Yeah. So good question. A few examples. So, you know, say, say an employee's access request, you know, fails or, you know, they don't get the access they need for for a particular tool, and they file a ticket. Right? Today, IT has to manually trace through the access policies to figure out which rule blocked it and determine, you know, whether the employee should have gotten access and then approve it or explain why not. And that investigation can take, you know, many of the folks you'll know, easily thirty minutes, maybe even an hour. But with AI that understands the environment, it can do that in seconds. Right? So it can say, hey. This request was blocked by this policy, and it shows you the policy. This is which requires manager approval from the employee's department head, for example. Right? And and maybe that approval hasn't been granted. That approval hasn't been granted yet. And so we're waiting that's what we're waiting on. So now the admin knows exactly what happened, and then they can decide what to do. Right? So you also don't want the AI to just go in and execute things without you knowing about it. So it understands the full environment, so it tells you, this is the reason why. Now, what do you wanna do? And then the action piece then links in right there, where you can either just say, No, I wanna grant it immediately, bypass the manager approval, or I'm gonna wait, you know, let's let the manager approve this and go by that common workflow. Or another example, let's take device compliance, right? Instead of running a report and filtering it, you know, folks here, I think, know how long it takes to build a report, and and manually figuring out which devices aren't compliant, like, let's say there's an outdated OS version, the AI can surface the specific issues like, hey, these 12 devices are running an outdated OS version. And by the way, these are employees that are maybe in the engineering team, and it's really important that they have an have an up to date OS version and then actually stage the remediation to, like, push the update or notify the employees and then log the action. Right? So the important thing is that the admin still makes the call. The AI does the investigation, and it lines up the fix. And then the human reviews it and approves it. So instead of IT spending most of their time on diagnosis and correlation between a lot of different data sets, they just jump straight to the decision. It should help you make decisions fast. I have always been a fan of the way we have rolled out approvals in Rippling and the idea that all of this operates with existing governance policies that you've put in place and approvals and permissions and and, like, things that you have built in Rippling already. I just keep thinking to, like, I know it's still because I haven't fully wrapped the future of it in, you know, fully fully in my head, but, like, how much a little tiny tasks and and fixes, like, people try and, like, hack together with, oh, how cool is it that I can, like, Slack a certain thing, like, lock computer? And it's like, that's so hacky compared to what we're talking about. So I just keep thinking about that kind of stuff too. So James, like, instead of spending hours diagnosing problems, the idea here that AI could essentially do the analysis, line up the fixes, perform most of the fixes. I mean, what are you thinking about in that context? I'm thinking about slacking my AI to fix the problem. No. So that that's the right way to think about it. The goal isn't to remove control. It's to remove all the manual analysis that leads up to a decision. Alright? So today, most of the time is spent figuring out what's wrong, and Zafra did kind of say that it was like, spent a half hour, sometimes an hour on these problems, but there's actually more time lost when like, real world, like, that's how much time it takes to solve it, but there's gaps. Like, often, there are silos for different software tools across an org where different admins run each. So that help desk tech, they take that ticket that's looking at the issue. It's not gonna ask someone else in the team. Hopefully, they're working a lunch, not hanging out on an island somewhere. And then when they get the response back, then they proceed. So it might be an SLA for that team might be four hours or next day. So now that one hour of work, it's actually spread over two days, and that's gone. Yeah. Right? Yeah. So if AI can do that part reliably, then the human stuff becomes reviewing and approving the action. Like, that's the part that matters, the approval and, like, the decision. You're still in control of the change. You're not doing all the investigation yourself. You're not paying people in three different teams, and it really starts to unlock real self-service. So I know that the hacky thing that you mentioned, the whole Slack, hey. Do this thing, but that's really what it will be like for the end user. The end user at your organization, once your rules and policies and governance is in place, provide me access to Figma. And then it's like, oh, well, they're on that team. There's an automatic approval process for people on that team. They provide the app. The link goes into your dashboard, and you get a password in your RPS. Like, everything just works because you sent a Slack over or or the equivalent thereof. So it's, like, it's there are many solutions that kind of pretend their way through it, but this is actually direct ties to system, which is which is great. It's not help desk deflection. It's giving users reliable answers and actual action without needing to step in every time. Yeah. And I'm certainly I don't wanna give folks the wrong impression about my my comment about, like, you know, having Slack triggers and stuff like that. I'm just thinking about, like, how often those could break in environments that I've supported before or how often, like, a change of ownership of this know, like, maybe Salesforce wasn't connected to my identity, and I was using something like that kind of hacked together solution. Like, anytime the Salesforce owner changed, if it's a HackTogether solution, I would have had to go change that myself. And that kind of stuff being naturally in the context and trust seems like it is the biggest, you know, the biggest coolest thing for me when I'm thinking through this. But we've talked a lot about the idea of all this, but it's it's obviously easier to see it all in action. So I think we're gonna switch gears for just a moment, and Zafar is gonna show people what this actually looks like in an IT environment. Yes, let's do it. It's gonna be a lot of fun. I am going to start with an across system operational question, so that you can see how the system traces through a live environment and handles this all in one place. Let me pull up rippling, and we'll walk through that scenario followed by a couple more. Just give me a moment here to share my screen. Okay. So I just typed who is starting next week? What apps are they set up for, and has their gear actually shipped yet? And the AI has responded with these results. This is the context piece that I mentioned earlier. Right? The AI is looking at shipping data, our MDM, and the hiring calendar all at once. I didn't have to message a recruiter or log in to a carrier portal. I asked one question, and I can see the full picture, who's starting, what they're set up for, and where their hardware is. I can see how many have completed onboarding, how many are in progress. I can see some key details. I can see the device order status for all these employees that are starting next week. If I want a more detailed report, I can access it. I can see the app access setup for all those folks that are starting, and then click into the reports as well. I also see how the system interpreted next week. So it gives me some sense. Both the reports that I can go and double click into and how the system interpreted what I asked, gives me some sense on the reliability of those results so I can trust what the system is telling me. Very, very cool. Such a cool example of operational this operational side of everything we're talking about. Selfishly, what about, like, visibility into, like, software stuff? Like, what kind stuff that might take you or folks looking for, like, software usage maybe a whole afternoon to to figure out. You got any any good examples for something like that? Yeah. That's a that's a great question. Let's let's talk about something that quietly drains every IT budget. I think folks will know what I'm talking about here, and that's license bloat. Right? So let's say I wanna understand where we have unused software licenses. In most environments, that means exporting a report from each app's admin console. Oh my gosh. Right? Then cross referencing it against your active head count. Right? And then manually building a picture of who's actually using what, even just talking about it makes me anxious. So so let's let's just ask Ripley. So I'm gonna type here, find anyone who hasn't opened Zoom in the last sixty days. Oh, and also flag employees who are on leave. So look at that. I just got a clear list of everyone who hasn't touched Zoom in sixty days. In any other system, I'd be inside Zoom's admin console, pulling usage reports and comparing that against HR data to figure out who's still active, who's changed roles, who left. That's easily a half day exercise to answer a simple question. And here, I have all the information in just a handful of seconds that I need to make a decision. Right? Now I can go and figure out, okay, like, I can go reclaim those licenses, downgrade accounts, reallocate seats to the new hire class that we just looked at. And then the point is that the AI gave me the full picture instantly, and so I'm not wasting any time assembling that manually. And this, like like, things you've shown already and things we've talked about are all stuff that kind of generate tickets, like like, historically. Right? What what about, like, day to day security tasks and operations that you you end up having to do? Yeah. That's right. Like, there is still the bread and butter, you know, huge volume of basic tickets that kill our productivity every day. And so it's not that these are unimportant. They're very important. They take a lot of ex ex extra time from the team. So let's say that an employee reports they've lost their laptop. This is how simple it is in rippling to make sure the data on their device stays secure. I can literally just say, I want to lock Erica's laptop. That's it. The AI disambiguated who I meant, it confirmed the right employee, and it surfaced the lock action directly. No back and forth on a ticket to confirm with Erica. No digging through inventory to find the device. The system already has the full context, and that's one less ticket for my team to touch. Crazy stuff. That's tying together like everything that we've talked about so far in this scenario. I mean, the AI has the context, the answer's trusted because it's pulling from actual systems of record, and the action happens inside of, like, governed workflows that you've already set up with, permissions that exist. So That's right. I can only access to Rippling AI what I am permissed to access. And and this is what it looks like when IT shifts from spending their day in a ticket queue to actually focusing on the architecture and security strategy that protects the business. The AI AI handles all the investigation and the routine operations and does the work. And then the IT team focuses on the decisions that actually matter. Super cool stuff. Well, I could make you generate, I could hope to see a lot more, but I think that's kind of wrapping it up for a bit today. We've been discussing a shift in how IT operations work wholesale. I mean, instead of manually analyzing the environment and executing fixes step by step, what's going on is Rippling AI can help teams understand what's happening and line up those actions to actually resolve those issues in real time. So, Zafr, again, thanks so much for joining us and giving folks the scoop on everything here. I imagine this has been a blast to see come together. Any closing notes or callouts to our attendees here? I I guess, like, the world is changing. Our jobs as we know it, they're changing, and it's happening faster than I think any of us anticipated. And that speed of change is only gonna accelerate over the next six to twelve months. It can be scary, but it's also extremely exciting. I think this is a a really interesting time to be alive. We are we are building for the future here, and we're so excited to help make your lives more efficient and more pleasant with the tools that we have just launched and will continue to to iterate and improve on. So awesome. Thank you so much for joining. James, it's been a pleasure as always. Any final thoughts on your end? Where could people see you next? Alright. Yeah. This has been great. It's all been a big shift as you're saying to this thing that is next. Stop looking for answers ourselves. It's like getting answers. And as Zafra was showing us, actually have it do a bunch of the work for us in ways we can be confident about. A lot of this day to day friction in IT goes away. That's the shift. And also see me next. You can play this video and see me here, but you can, of course, find me on LinkedIn. I'm also on Mac admin Slack, Jay Sorrenti. There are a few more different conversations I'll be on on the calendar and bunch of in person events coming up. So I'll be here to keep you all updated in the spaces I hang out in, and I'll see each of you out there. Thank you so much. Alright, everyone. That's a wrap. Thank you for joining us today. Have a super great rest of your day. Cheers. Hello. Hello. Not done quite yet today. Thank you very much all for joining us. I hope you found that, helpful. It's been a exciting couple weeks, around here at at Rippling and Rippling IT in specific. So, yeah, thank you again for for taking the time as always. Love to see some repeat guests here as well. So welcome back to anyone who's ever come and hung out with us before. James will be back in a sec to there we are. Hello, my friend. To handle some of the the q and a that came in. But, like, final call for any other questions that that might be on your mind about what you just saw, what we talked about, or, you know, whatever else happens to be on your mind. We will do our best to get you an answer. Real quick, I am just gonna run through a couple other webinars that we have coming up. Love to see you guys back here again. Same time next week, we're gonna be talking about compliance would be the short version of it, but kind of misconceptions around compliance and, you know, how those systems can be engineered for the the highest degree of success. And we have a a really fun and and cool guest named Keith Badri, multiple time CISO, a very experienced security leader who just he has a a I think, a very well refined perspective on this and, you know, in conversation with our with our two our two standbys in in James and Carter, who you just heard from a few moments ago. So we'll be back for that next week. That's a really fun one. If that's something you're interested in, please stop on by for that. The following week, same time, webinar o'clock, twelve p, Pacific and and 3PM eastern. We'll be, with one of our customers named Nella Juma from AC Disaster Consulting. And, yeah, we're just gonna talk about, like, her experience moving off of an MSP and onto Rippling and, like, where she has found that has really helped her, you know, streamline all of the all of the work that she's responsible for, from a technology standpoint. So, like, that was that's gonna be a really great conversation too. So so please take the time to to join us for that one, more of, like, a case study flavored type of convo that day. And then a few weeks after that, we'll be sitting down with yet another customer of ours, a New York based one, which James and I have spent a lot of time with, but very excited to to sit down with our good friend, Josh Mullis, from Productive. So that will be on the twenty ninth. And, again, that's gonna be kind of how productive and Josh have just managed to to really eliminate a lot of the, frankly, annoying ClickOps that used to take up much of this day, week, and month in a former pre rippling life. So, Josh is great. I think you'll really enjoy hearing from him. So please come on down for any and all of those sessions. But, we will get to the, raffle winner in just a moment, but I do wanna make sure that we are taking a stab at some of this q and a. So, James, if it's cool with you, I'm just gonna start teeing you up with a couple of the ones that I see coming through here. So let's start with, I'm I may go out of order here for which I apologize. But, David, during I'm guessing this was during the the demo session. Question. The device ordering status, is that serviceable only if we use Rippling for device orders? Alright. I'm gonna answer this question. And if I'm incorrect, I will definitely follow-up later. Yeah. So worries. device orders has only serviceable when order is tracked in Rippling, but it does not mean they haven't been purchased through Rippling. If you're using our device, our warehousing service and we're we're doing that work, we have those workflows, that status will be will be shown, but does not need to be purchased through us. Great. Number two, this one is from Cesar. How do you handle approvals for services that are not on SSL and require native accounts created directly on the SP portal? This one's even got a thumbs up from him. Oh, I just added one. Sorry. From yourself. Alright. So Michael. loves this question. right. I do. Yeah. I guess it depends on a couple of different scenarios. Every app is a bit different. There are manual access administration for certain ones. Some of the might be a process of SSO flows aren't available. It might have SAML. It might not. So, essentially, it really depends on a specific app and what the process is for it, but we do have processes that work outside of direct integrations. I, started a trend. the question. Fair Fair it. hand. waving? But. yeah. Another one from Cesar. Does Rippling plug in directly to the Apple ecommerce site to place laptop orders? Two part answer. It doesn't plug into the ecom oh, three part answer. It does not plug directly into the ecommerce site, but it does work with Apple Biz Manager, of course. So if you are using Apple ecommerce, things will feed into there. But more importantly, we're also an authorized Apple reseller. So you would place orders directly through our device store, and that's available no matter which, to whatever SKUs you have, you can use the device store, and those will automatically be registered through Apple authorized reseller and go into your Apple Business Manager or Apple Business as soon be called, and work with your automated device enrollment. Right on. Very important question here. I don't know if you've already seen this one, James, but this is from Tyler, and he has asked, James, how fast can you solve that Rubik's cube in the background? Winky face emoji. This guy, about fifty seconds is my official current score. I'm not gonna do it live right now because just about to say, you wouldn't dare put your money where your mouth is on that right now, would you? no. Maybe maybe later, but find me find find me elsewhere. Will 100% perform for the crowd. Very well. I do not wanna I do not wanna put you in that position. Cool. I'll save the best for last. There was one that came in during the session from Antonio asking about API integrations. We were just messaging about this back and forth. What do we have on that one? Alright. So I I saw the message in the chat. I I couldn't know, I was trying to also talk, but I saw it. Honestly, I'm not a 100% sure the depth of the question. How question was, how does Rippling handle API integration across the board? What about legacy? It might have been a direct reference to what was being talked about that segment, but I did, drop it into ripple.ai just to play along with the exercise. And it gave us it did a couple of things that sound very interesting. It broke it down. Right? So it said how it actually works. When you have an API subscription, you can generate API token through tool developer API tokens. That's basic stuff. Legacy API shouldn't be used for new integrations, but, like, across the board, it, you know, it works for custom builds and app based integrations through the app shop. So my point is there's a whole lot of things that could mean and could go into. But when I popped it into ReplyAI, it gave me a detailed part of each section about regular about across the board API and legacy, and then gave me direct links to the actual articles that talk about it. Hopefully, you can get to your answer that way. Also, it is a more specific question, but, like, a a more framed question with context. I'll try and get it answered for you. Email me or find me on, like, LinkedIn or wherever. The message that we sent out after this thing. I'll get you an answer. But I didn't have enough context to give more specific answer. Right on. Well, hope that was helpful to all who asked questions, and thank you, James, for for sticking around to do that with a brief drum roll. I believe that unless there are any last minute questions, but I don't see any. So I am going to jump into our beloved raffle here. This week's is fun, and I would say, relevant to the topic at hand. I know James owns a pair of these himself, which is very true to James. There there they go. So some second gen Ray Ban Meta Wayfarer AI glasses. I think I caught all the product text there. These are pretty cool. I actually don't know. I don't have these, James. You can tell me later if they're cool. But I do hope that Mark Warkowski from Blue Cross of Idaho thinks they are cool as he'll be the proud owner of these classes from this day forth. So there we go. See, he's excited. Cool. We will we'll be in touch about those. I think tomorrow usually is how we how we do that, but look out for another message from us. We will make sure those those make their way to you quite soon. So with that, I think we can wrap it up. But wanted to thank everybody once again for joining us today. I always appreciate you spending your afternoons with us. So much appreciated, and we'll see you again soon. Take it easy. Thank you all.