Transcript
Okay. Today's topic is Autonomous IT Masterclass Robin. It's the part one of the three-part series. We did a similar series in the past quarter. This time, it's a little bit more in-depth. We are getting into the weeds of how to configure AI in 2026. If you're new to the product, to the offering in the AI field that we have at Eterra, I'm glad you joined. Today, I'm going to walk you through the foundations of what we do offer, how it all works, what is the most important to be aware of, and slightly less important, but still important things to be aware of, and how to build a successful AI-based IT system in 2026. Okay. A couple of words about myself. I'm Dima. I'm the Technical Product Marketing Manager at Eterra. I used to be in professional services and was responsible for implementing Robin and Eterra's platform for the past almost three years, and now I joined the marketing team as a technical person to do such sessions, so you get more details about the most interesting and most complex parts of the product, even though our setup is relatively simple. However, we have some things to go through to make sure you succeed with our solution. The housekeeping rules are super simple for today's session, so yes, you're going to get the recording in the next 24 hours after the session ends. As I mentioned before, you will see the Q&A section right next to the chat icon, and you can post your questions there, and to confirm that I see them, Gypsy, yeah, so can you get a copy of this? Definitely, and let's take a quick sneak peek of the agenda, so introduction, what is Robin? Key things to know when you are touching this topic, what you need to consider before implementing an AI solution. In 2026, the demo, which will consist of two parts, we will go through certain settings section, sections which are worth mentioning before we even begin to work with the product, and then we'll walk, and I walk you through through a couple of use cases that are, let's say, level one, as I may say, the most simple one, but yet one of the most common ones that we see across the customers who already implement Robin as we speak. Okay, let's start from the very beginning. What is Robin, and what is Atera in general? So in case you just joined us, if you're new to the platform, new to Atera as a company, and you're not an existing customer, hey, Don, good to see you, it's been a while. Okay, so Atera is an IT platform which consists of RMM and PSA components. We target patch management. We oversee the network discovery. We have the AI co-pilot as the technician-facing, AI-powered solution, and we also recently introduced the asset management, and now we are introducing Robin as the next step for the end users so they can get the same benefits as technicians do from the technology that is all the rage right now and is all over the place. So what does Robin do? So Robin basically targets your tier one tickets and acts as a tier one technician. The one that requires the same level of onboarding as a real person, the one that needs a little bit of knowledge about what your company does, how you operate, the software that you're using, and as long as the context is built well, it's going to operate in your ecosystem and will be perfectly aware of its surroundings. What do I mean by that? And how Robin integrates into your system. So for starters, yes, it's a 24-7 digital tier one technician which lives in our customer portal or service portal, depending on the type of the product you are on. I mean the IT department or MSP respectively, and it does several things. First of all, it knows the devices that belong to users who talk to it, and it assesses before responding. It assesses the device context to get real-time device telemetry and diagnostics. It not only gives advice, but also it acts and does things on the device, and it all thanks to the amount of context and the presence of the Atera agent, and this is one of the very important pieces of context that it needs to know, which device belongs to which person, and for you as an IT person who is managing the tickets on the other end, correct user assigned to a correct device means you have the right context to work on, and for Robin, it means it can do the diagnostics which are relevant for this specific device and for this specific person. Robin is trained on the basic IT knowledge, and it knows when to throw in a towel and call for help, and by help, I mean escalating to your IT team, and if this is something that is, for instance, like hardware-related, so I mean like getting approval for a pair of headphones or something like along those lines, this is where it will definitely escalate because it cannot hand you over a pair of those, but still, and also it also introduces you as a company to a new type of habit which is more relevant to 24 to 26. Those years are when AI starts to be all over the place, and it evolves together with you. Why do I mention that? Because during those two years, you are, and even before, you started to accumulate your company context, meaning which is like knowledge-based articles, scripts, anything that is relevant to your organization, and it definitely will help, and what do I mean how it can help and evolve with your tech stack and how it activates it and lives in it? First off, as soon as Robin starts to exist in your environment, it has many entry points from which it can take over and address end-users' IT and not-so-IT requests. You probably have seen like a bunch of familiar logos which I just shared on screen. It's Microsoft Teams. Whoever's not familiar, Microsoft Teams. It's email, WhatsApp, Slack, and natively on the portal, which is web browser-based, so this is where we, this is our promise to meet end-users where they are. We are thinking of more omni-channel capabilities, but this is definitely a solid start, so this is what we, this is the type of knowledge we acquired while we were doing the market research and starting with Robin. Next, the next layer that helps you to create a solid stack which creates a nice blend of AI and your, what you already have. Those are APIs, automations, which are powered by Zapier, and Cloud Actions, about which I will talk a little bit further down the line, but in a nutshell, those not just APIs that are built into Eterra, the Cloud Actions are something that is, are something that is designed to help you talk to and help your end-users talk to any third-party systems of your choice and retrieve relevant information for the tools that don't have any type of native integration to Eterra, with Eterra, and, but still, you need this information to succeed with your day-to-day work. And the last but not least, but super important, the deep intelligence layer. Step one, and the layer one, is your knowledge base. It can live natively inside Eterra. It can live outside Eterra in solutions like Notion, solutions that are provided by a company called Atlassian and Confluence. We also plan to integrate with more knowledge-based sources. And the deep intelligence that I just mentioned, it's not only powered by your, let's say, company-specific knowledge. It also allows you to use the MCP integrations. For, to anyone who is new to this acronym, the MCP stands for Model Context Protocol, and this is basically a technology that started to emerge alongside with the AI tech. And if you're using any of the AI tools out there already, it allows you to establish a full end-to-end, two-way conversation with a third-party tool that has an MCP connector. And it allows you not only retrieve, and retrieve and post information there, it also allows you to do things. For instance, if we're talking about like Notion or Confluence, it allows you to create knowledge-based articles in a certain style. If we're talking about tools like Asana or monday.com, this allows you to create projects and manage projects there, and also follow the updates on the ones that you're mostly interested in. And the list will eventually grow, and this is what we do have right now, because custom MCPs is something that we're working on, and we're working on many other exciting integrations that will help you to make the decision to switch towards the AI-powered autonomous IT. Okay, without further ado, let me switch to the product itself, and tell you about certain things that I found the most important about configuring Robin as a feature. Because Robin, besides the fact that it's an artificial intelligence-based technology, it also needs something before it can fully operate. And the most important thing that I, if I would say just one word that I want you to take away from today's session, the most important part when you're implementing an AI solution like Robin is context. Context is king in all AI solutions out there, including us, and probably you have noticed when you're just asking a very generic question, and it gives you, like any AI tool will give you a generic response. However, over time it learns how you talk, how you write, what you do, and this gives it enough background to work more efficiently. With that being said, let me switch to the product and show you what I mean, okay. Okay, so for Robin's configuration starts with something that apparently if you're an existing Atera customer, you have the step zero done. And the step zero is establishing in a solid Atera agent base across your end users machines. So if we are talking about like an organization of a thousand devices, it means that a thousand of those people have a Atera agent installed, and they're working on their laptops day in, day out, and they have the Atera agent already there. So step number, like it's still step zero. It's part of the zero step is associating the end users with the devices that they work with and work on. Why is that super important? Because the contact data that can be synced with your Entra ID instance is a very crucial when it comes to a context of where decision-making for AI, because it needs to know which device is it talking to, which user is it talking to, and which user is on this specific device. And from there, it will provide a more context to where diagnostics, as I said, and it will know how to act with this or that person, depending on which, I don't know, department this person belongs to, which title this person has, and many other context-based things, which it cannot read off, like it doesn't have any type of Jedi mind tricks, so to speak. So it needs context to do meaningful things, and I would put emphasis on the word do. Okay, the first page that I want to talk about today is that lives in the AI Center. As soon as you get access to Robin, as well as the IT Co-Pilot, which is out of scope for today's discussion, but still somewhat relevant. but still. So the AI Center is your HQ for every single feature that is AI-related. The very, very basic configuration is relatively easy. You can connect your corporate messengers to Robin, then you can establish the language that we'll respond in. And as you can see, that we introduced several features recently that were not there even a couple months ago, one of which is Proactive Assistance, which allows to, when a user is engaging through an email, the old school way, so to speak, it allows Robin to send a link to the portal so the AI experience in the browser, in the chat window becomes more fluid because some people prefer emails over anything else. For the ones who connected, their corporate messengers is way easier because Slack and Teams, as soon as they are connected, each and every end user can start chatting, throwing in attachments, asking away the questions that they might be interested in, and all of that nature. Next up is the ticketing aspect. I will zoom in to make it more reader-friendly. So first of all is that in this section, one of the key things is that you need to let people know that Robin is becoming their first stop when they want something to be done by the IT team. And as you can see that the toggle says exactly this, create tickets via IT autopilot only. This means that if an end user will try to create a ticket using the old methods, they won't be able to, and they will be redirected towards having the Robin-based experience. In parallel with starting a new chat with Robin, tickets are created in parallel, they land in the same ticket section inside of Terra, and like IT autopilot, it's not Robin in the UI, but the naming change will come in the coming days. Assign Robin to tickets created by email, so it will be assigned to all the tickets that are coming the old school way, and it kind of complements creating the tickets using Robin only. And a very new feature that has just landed, it will allow Robin to make the human escalation experience more fluid and make it flow better. It's filling in the mandatory custom fields, which just reduces the time to resolution, and it caters all your needs when it comes to the data that you're having to maintain and present to your upper management whenever requested. The next section is super important, and it's a fine-tuning of how AI as a technology should operate with an end user, and that is follow-up and resolution. I will start from the bottom. Wait time to follow-up. So basically this setting is responsible for how often should Robin send nudges to an end user, depending on the channel that they are approaching Robin. So the chat means everything but the email, and the email speaks for itself. The second one is wait time for the next action, like if a person is not doing anything, is something that is dedicated for how long should Robin be on standby before he can move on, and close the ticket because the person stopped responding, and we've all been there because just, people open IT requests, and some of them just say thank you, which is a very popular last message on the ticket. Some of them just leave the window open and just calling it a day, and the ticket stays open, and this inflates your time to resolution metrics for no reason, and this is dedicated exactly to this. The last but not least on this page is the resolution policy. It's one of the most recently added features, and it has two behavioral settings. One is conservative, the second one is optimistic. So the optimistic allows Robin to operate more freely in regards to how and when should any of the IT tickets should be considered as resolved, and conservative means that the end user should explicitly mention that the ticket was actually resolved so Robin can go ahead and help you close the ticket. Next portion of super important things that I want you to be aware of. So number one is definitions. Definitions is where all your company-related knowledge lives so it contains several sections here, one of which is a company description. I will unfold it. So basically it's a short description of a company that you work for, what it does for a living, which branches doesn't have, any type of, anything that you find relevant in up to 600 characters. It's done for a reason to prevent the context bloat and keep the description super focused on what it needs to do. And then right down below, you will see the actual definition. So the definitions consist of two parts. Number one is software. So basically those are the software items that you have in your company as your technology stack because if you put the context for software here, so every time a person asks a question that is VPN related, Robin will always know it as OpenVPN. And if a person is mentioning something like DM, which is direct messages, can be something else in our company's context, it's the actual pay-per-company that we're talking about today. And yes, it's a nod to the TV series, The Office, Dunder Mifflin is the one. And as soon as you enable it, the system will guide you how to write it all down. And it will allow Robin to have more educated response to any type of questions, even with a very vague wording. Because users usually type in a very natural language. Not all people are technical, not all people are good at prompting, which is a nice skill to have in 2024, but in my humble opinion, a very important skill to have in 2026. And being able to ask properly an AI product, something that they need from it, the way you ask, the way you prompt, defines the quality of the response you're getting in return. And it will affect the overall experience, and it will affect the outcome and how fast AI can act before it escalates to a person if needed, or not at all, and it will be automatically resolved and you can leave happily ever after. Next, something to configure for users to consider, you can also add not only the description of the software that you have out there, but also you can add the software that users will have as a whitelisted list. And they can install this software with no problem, but also Robin will be aware that they're all allowed to use the software and without any escalations or anything, they just can install and have it and use it right away. Next, super important, probably the crossroad of everything that is context related. The place where you should be describing how, like in which order, how, in which parts of the context should matter, and be, so AI can act on proper things in proper order. For instance, like I have a bunch of playbooks right here. The playbooks is basically like a guided step-by-step instruction for Robin in which step, which action should be taken, and what is needed to be considered when Robin is acting on a specific task. In that specific case, like software versus job role. So what does it mean? So if a person belongs to sales, check if they're asking, like if I have all the software installed in my laptop, it will check only for software items, Zoom, Chrome, OpenVPN, and Slack. But if a person belongs to product marketing or design, it will check way more solutions. It will also check Miro, it will check Notion, and Figma Desktop, which is especially relevant when it comes down to teams like product, design, and anyone in between. And as you can see, it can include different guidance steps. And as you can see, it's not directly editable for security reasons to prevent a prompt injection. So, but here you can chat with it in free language what you need, your scope, your ask, like how you spec this step-by-step guide. And as soon as you enable it, in about 10 seconds, and it becomes available, and you can start doing things. And whenever you ask a question, it will follow exactly these steps. Another thing that I would like to mention is something that we are working on and will be out very soon, which is also important for the company context before I move to the actual use cases. We will have a report which is called AI Readiness. What does it mean? It will give you an understanding of how your company looks like. I will show you the preview of this report. This is something that we will show you really, really soon. So basically, the AI Readiness report will, A, assess the total number of agents you have in your environment. It will give a definition and count all the AI-ready agents, meaning that you have an agent-based device, which can have a Tera Agent installed on it, that email is attached to a contact, and it is attached to a device, and this contact also has a mobile phone. Why does mobile phones matter? So mobile phones matter, make or break, any type of experience for sensitive actions, such as local password resets. It will be one of the use cases that I will be demonstrating today, because if you're locked out of any, out of all the systems, or your laptop is blocked after three unsuccessful password entry attempts, your phone will be used to send you the link to supply you with your temp password, and in return, get you the laptop access back in a matter of minutes. And as soon as the report is ready, I won't show you what's down below, but in more details, but in a nutshell, this will affect the data that you will see in this report. It will define how easy it will be for you to implement an AI solution, and how healthy your environment looks like, and it's valid for both Robin customers that we have, for customers who have Robin and AI Copilot, or just AI Copilot, because AI Copilot will also be super useful for the people who are, so the AI Readiness Report will also be useful for the ones who are doing AI Copilot, powered actions in the environment, and this is something that we'll probably talk in one of the next sessions. Okay, so let's start, and before we go to actual demos, so the very last stop before I move forward, super important, knowledge-based articles, and in the AI Center, you will also see the AI Insights. Those two things matter, because all your successfully resolved tickets turn into knowledge-based article drafts. Knowledge-based articles are the context on how Robin and AI Copilot should operate, and what they should be aware of, you know, whenever a person is asking this or that question. Not only that, is that if you are having a remote session that is a repeatable action, it's also valid for both AI Copilot and Robin. If you're using Splashtop, your remote sessions will be assessed on a script level, it's not capturing the screen, it will be on the script level, and like you did, two remote sessions, which could have been a script, and it can automate yet another 10 to 15-minute session, saving you time and you know it very quickly builds up over time. Now let's switch gears a bit saving you time, and you know it very quickly builds up over time. Now, and see what it can do. Let me zoom into my laptop which I have standing right next to me. So the Robin experience starts with people that are being welcomed by the welcome tour on the portal. It's going to get some upgrades real soon but what it offers today, it gives an understanding to an end user how it all looks and feels, what it can do, what can a person get help outside of business hours, and when is it escalating to the team. But most importantly, the context-based layer is when it diagnoses your laptop and in return gives you the device-based data. And this last step is crucial whenever it's finishing its diagnostics. It will show you the hardware data like what's the CPU, what's the RAM amount, what's the free space, the serial number, everything that it knows about a device. And as soon as it does it, it will say that we're good to go. I will probably skip this for now, but I wanted to show you that it knows that it's working on the test laptop and I want us to start a new conversation. So the first use case I want to demonstrate is something that is pure knowledge-based driven and that is like I'm going to a trade fair, what do I need to prep? I'm trying to make typos and make it as hard as possible for the AI. So as you can see, the ticket has been submitted and immediately gets the ticket ID and the thought process has started. So basically, whenever you are traveling for business, and this is not a pure IT use case, however, whenever anyone is traveling for business, it's a huge stress. It's one of the questions that live under the carpet and people are afraid to ask because they're afraid to sound incompetent. But apparently, Robin showed us over time that it helps anyone in the company to feel the safe space where there are no silly questions to be asked and it immediately guides them to relevant KB articles or gives them the top level experience, top level description of what they need to prep. For instance, can you check my laptop if it's ready for the trip? So this is one step further. Knowledge-based articles are nice, however, it also needs to know if your hardware is good to go, if you have every piece of software that is relevant for business trips ready. And also, as you can see, it can double checks if you're taking, let's say, the expo laptop versus you're taking your usual work laptop. Is this laptop you'll be taking on the trip? Yeah. And as soon as I do this, it goes back into the thinking process. And as you can see, I'm checking the laptop details right now. I'll verify if it's ready for battery health. It will check if it's good to go in regards to VPN readiness. Everything that is, like, if there is enough free space on your hard drive and everything in between. It will take some time because it gets through several actions in a row. And the diagnostics should end up in a minute or so. And while we wait, let me switch back to my Atera UI to show you which laptop that I'm talking about. The laptop that we're testing today is this laptop. So this is a Lenovo laptop that I'm working on. And as you can see, immediately as I started to do the screen the hardware. And this is the laptop that I'm talking to. And this is like, and as you can see that already I have some actions listed here. And then it allows me to get all the necessary information. So it knows all the context. Oops, sorry. Back to our remote session. Yeah. So as you can see, while I was showing you the user interface, it was like it got the BIOS information. It's looking for actions of the WMI query. So everything that is required to do your work while being on remote. And as I said, it's not doing it super fast for the reason that it's going into the weeds of the diagnostics. And as soon as it's done, it will give you the verdict that your laptop is totally fine and you're ready to go on a trip. It should happen any minute. Yeah. And as you can see, the reason why it's doing this is that the playbook that triggered the trade fair readiness, it triggers so many diagnostics and it needs to make sure that you're not getting into trouble, especially if you're presenting on the trade fair and or maybe you're having a very important meeting or you're working in a company that without, for instance, like a VPN client working properly, you cannot do much. So let all the diagnostics finish. They do take time. And while they're here, let me switch back to the product. So I wanted to show you something while we wait not to waste our time in the reports in general. You will see the recent processes. And as you can see, I started like the diagnostics just now and it's doing the WMI query search. So everything is traceable. And as you can see, this is the model manufacturer, the return of the script that is running in the background, get graphics adapter status so I can check the output of those scripts. And as you can see, it's doing an extensive work when it comes to hardware diagnostics and most importantly, software diagnostics. So for instance, like top five running apps, as you can see, it's like Windows Explorer, Microsoft Windows operating system show OSD and things like that. Let us get back to it. Yeah. So as you can see, your processor information, disk storage, battery graphics, everything is checked. And this is what an IT person would do before someone goes on a business trip. And as soon as it's done, it gives you the verdict that your laptop is good to go. And as we're getting more and more check marks, it gives us an understanding that we are actually in a good spot. Okay. It should be done any minute. And as I said, for security reasons, one of the huge concerns and like huge questions that we're getting, very often questions that we're getting is like how you can trace what it did on the machine. So in the recent processes, thanks to the new UI and don't you were in the yesterday session. So this is something that you can definitely attract and trace. Let's let it finish. As soon as it's done, we will continue from here. And as you can see, it did a very, very extensive check. Which is a good thing at the end of the day, because it's doing its best to not let you fail on a trade fair and not to be locked out like five minutes before your meeting. Okay. So I'll verify that VPN piece, physical disk health. As you can see, like it's end-to-end diagnostics without even leaving your chair. We'll return in a second to save us a couple minutes. And I will show you the end result of this diagnostics. I wanted to show you an extra use case while we wait, which is something that is super, let's say happens all the time. And people are joining online meetings all day, every day. It can be jumps in between the Zoom or Teams or both. And sometimes, can you hear me now problem occurs. So can Robin assist you in resolving those type of issues? Because usually people are rushing into a meeting without checking their headset. If it's like a Bluetooth device, if it's a USB device or something else. If it's like a separate mic, if they have a sound interface. So as you can see right now what's happening on the screen. So I have a USB headset. It's loading a playbook and it's been like now it will like it will load the playbook. Is your USB headset is connected? Yep, it is. And right after this, Nelson, the answer to your question is like your AI center looks short probably because the Robin is not enabled. And maybe you just have co-pilot. And this might be the reason why. And while we're watching it, so let's like the mouse cursor on the screen will move any second next to the taskbar. And around this area, you will see that it will ask for an approval to switch something in your hardware and hardware values. And before this, it needs an approval. It goes back into thinking process. And after like 10, 15 seconds, it was like approved. And I'm applying the headset audio switch now. And let's take a look at the right side of the screen. And I didn't touch my screen. My hands are up here. And I changed it from speakers to headset and earphone and headset microphone. So we're good to go to join a meeting in under 120 seconds, which is exactly the promise that we have. Let us get back to our diagnostics. Yeah, completed successfully. So it's about to finish. And let's watch one more use case that is super common, which is local device password reset. Starting about now. So basically, if a person gets, this is why the mobile phone is important. It all starts with the wrong password, which is pasted here three times in a row incorrectly. And right after this, the screen capture will transition into the phone screen, which will demonstrate what happens on the end user's mobile phone and why is it critical to have it. And as soon as it's done, you will get a message, which is an SMS message. It's a Robin will say, hey, Michael, this is my stage name. And it will tell you like, here's your link to your password. You'll click on the get new password. Right after you do this, you should wait for several seconds and get your temporary password. Get back to your laptop, paste your like your password. I was pasting it because it was a remote session and you're back online. And yeah, it all thanks to a terror agent installed on the machine, which is definitely something that is super good. Okay. While we were waiting, as you can see, Robin went through heaps and heaps of diagnostics. But the end result is pre-travel check complete. Your laptop is ready to go. Here's the hardware. This is what is installed. I just put one hardware concern. The current terminal reading is high while battery is okay. And the hardware concern, like the reading is high because I intentionally put a small piece of cloth under the fan just to emulate that the temperature can spike and it will notice it in almost real time. And that basically concludes the demonstration part. And let's switch back to the takeaways for today's session. So what makes Robin unique? And right after this, we'll switch to Q&A. So Robin is something that we're offering in the realm of autonomous IT for the modern enterprise. Customers that do it right, have enough context and collaborate well with, it's not just the IT department realm. It's just a habit rebuilding inside our organization to implement a new piece of tech, which is different from any SaaS product you ever implemented. AI implementation is something that comes in loops, in iterations, and it requires constant attention to details. So it gives you proper results. Zero, like one 10th of a second, first response time is our second promise. And as you have seen with the audio interface And as you have seen with the audio interface example, example. exactly 120 seconds resolution time end to end for most of those cases, which require not so complicated and tedious diagnostics, which is like relatively simple one or two or maybe even three step execution of the actions on the device, which are possible thanks to the presence, as I said, that agent and they act as like any other agent basically allows the elevated access slash admin access to exist in order to execute things on a device and do things in general. And yeah, this is what we think of the ID tickets. And we want you to do meaningful things at work and say goodbye to those type of tickets forever. And this is like the good intentions only. And let's switch to Q&A. One of the questions that I got while I was presenting was a question from Maxwell. What's the difference between Robin and Co-Pilot? The Robin is, let me switch several slides back. So it's in front of our eyes. Robin is end user facing AI offering that we do have. It's a paid add-on, which can be discovered with our pre-sales team and our sales team. They will tell you about that in more details. AI Co-Pilot is something that is technician facing and end users have nothing to do with it because AI Co-Pilot is designed to do the ticket summaries, diagnostics for the technicians. For instance, like instead of remoting into a device, AI Co-Pilot can help a technician to automate the script injection or to help a user out without needing to remote into this device and make like hardware diagnostics without even looking at the device to begin with. So this is the core difference. So two different audiences, two different products. It's like living in the same apartment but in different rooms. Nelson, how do I enable Robin? So first stop, your first stop will probably be to request a demo. It's in the top right-hand side. If you are an existing or a trialing customer, we can enable a Robin demo for you. However, it requires a certain bits and pieces of setup before you start using the solution. And this is why a little bit of guidance will pay off in a long-term implementation. Also, Nelson, you said like my AI center looks short and don't have features like yours. I responded to that question. Don, will it be able to answer questions on Microsoft Office or Google Docs? Not sure I follow the question in its entirety. Do you mean that it needs to retrieve the information from like a cloud-based document or will it be able to respond to an answer about those specific software products? So you can paste it in the chat so we can answer it quickly. And also, Don, also a question from you. If we don't have a database to feed Robin in advance, does Eterra offer general best practice articles that we can add and edit? So this is on a case-by-case scenario, but your historical resolved tickets will be an amazing foundation because we can index them and we can find an inspiration to create KB articles that will be the foundation of your future Robin's brain. Okay, if the client is asking a question about using MS Office and Google Docs, it has general knowledge about both of those products. If you have something that is very specific and to your organization, you need to paste it as a knowledge-based article. And for instance, like how do I open a DocX document? And you can add it as a definition to give it a context. So the DocX documents are open through Google Docs because in the company, like I don't know, in the company ABC, we're not using MS Office to begin with, we're using the G Suite. And hence, the Google Docs is the go-to to open a document there. You need to do step one, two, and three, meaning like uploading it to the drive and open it with Google Docs. And move on from there. But the general one, like SaaS products 101, if they're out in public, sure, it will give you the knowledge. With MCP servers, it gets even better. It will give you a very extensive answer. It can actually become a sort of, let's say an enablement, a close to enablement product. So for instance, imagine you have certain softwares like Asana or Jira, which are, let's say, very complicated products and they have many, many layers to them. And you can basically, using the Robin, teach any person in the company how to use those products. And essentially, anyone who has access to this specific product becomes the product expert immediately because all the product knowledge, which is fetched in real time through Robin's system prompts, it becomes available and it allows people to do things because apparently the industry says that about 28% of the time that people spend at their workplaces is wasted on searching for information and not doing things. And this is what we want to eliminate and make everyone an expert in what they have in front of their hands in terms of tech. How, Alfredo, so how does it integrate with other platforms previously listed? Let me switch back to the MyScreen sharing. It all lives in the AI Center. Integrations live here. By design, it has integration with Microsoft 365 Lantra, Google Workspace, and Okta. It also has an integration with the MX Toolbox for whoever is interested for DNS records purposes. As you can see, we are now offering six built-in NCP servers from the solutions that are listed here. Custom MCP servers is something that you need to host in your end and configure yourself. However, our professional services can guide you through the process. It's not something that is sophisticated. And to configure cloud actions, for instance, like you have a network tool maybe that you need to fetch information from. It's more for technicians, but I'm just thinking off the top of my head. If you need to fetch any data that can be used in any type of report, why not fetch it from a third-party tool, which they might not have an end-to-end integration with the Terra. However, whenever you do a new cloud action, you can either create a REST API request yourself, which we'll touch on the next session, or generate it with Copilot. Just be descriptive, know what you're looking for, add an API key, and it will move from there. And we have the time for yet another question. So in the MSP platform, how do you define the software used by each client? So the software. So you can have a pool of software that is available to everyone. However, you can also instruct with Playbooks, which software is not available to this or that customer. On top of it, whenever you are creating software bundles, you can define to which IT automation profile it is assigned to. So the expectation is to have the software assigned through the IT automation profiles, and with very little exceptions, to install it for IT automation profiles, which can be customer-specific. That's the quick answer. Okay, I think I've responded to all the questions, and that will be probably the last stop for today. If you enjoyed, sorry, I'll be quick. If you enjoyed today's session, or enjoyed it in general, just if you're an existing customer, just leave a review on G2, which we definitely encourage you to do. And yeah, we'll meet each other in a couple of weeks from now to uncover more advanced use cases for Robin, and how it can elevate your workflows, and how it can help you solve things that you usually waste time on, and instead of doing meaningful things that IT people should be doing. Oh, before I go, I just noticed two more questions. Sorry, they were not in the Q&A. Can the agent replace Azure AD and Microsoft Intune for logging into Windows? Not sure I follow the question, Shlomo. So, it integrates with Entra and Google Workspace for logging in, but it doesn't replace it. A Tera agent complements it, and then allows all those things to talk to one another. And for instance, integration with Entra ID, especially when you do the Entra ID sync as an enterprise business app, it will allow the cloud passwords to be reset, and it's one of the superpowers that Robin is capable of. It will be capable of resetting the password for anyone, but Entra ID admins, that's the Microsoft security. Red line, and this is what we cannot bypass. Can Entra AI provide network diagnostics granularity as well? So, this is something that is not, let's say, lives in the Robin's realm. It's more of the future of AI Copilot. And I see some names from the yesterday session, and this is something that will be introduced in the coming quarters. It's the AI Copilot action mode, which will allow you to do the environment-wide queries to get actionable data at scale. And this is something that will definitely benefit your workflows and save you some time. We're also thinking of clever ways of integrating AI into our reporting component, which will just give you the data you looked for as fast as possible on top of the looker that is already integrated into the platform. I have no ETAs just yet, just sharing what I know as a product marketing manager. Okay, last check on the questions. And if no more questions, all I have to do is say, thank you, stop sharing my screen. And tell you that you're more than welcome to join us on the upcoming webinars. Everyone will get the recording, as I said, of this session as soon as it's available to you, is available to us in the cloud, and you will get the link to rewatch it. It will be available on demand. And we'll see you on the next one to dive deeper into more advanced use cases. And on that wonderful note, thanks for joining me today. I hope it was something that was useful. You learned something new. And thank you so much. Bye-bye, see you on the next one.
