We are on time and thank you very much for everyone who joined us today in this webinar series, where we’d like to start our discussion about a recent topic about how generative AI revolutionizes case management. And we prepared the demonstration and couple of slides to you just to show that how Eccentex is approaching this very exciting opportunity, how we are able to do something new. My name is Tibor Vass CMO of Eccentex. I will do it this presentation together with my colleagues Maksim Gill who is the Vice President of Product and he prepared a demonstration where we actually will show how this technology can be applied into the day-to-day business. So without any further a due. I would like to just start my presentation.

With that slide which is important to define Eccentex, how Eccentex is today.

Probably most of you already know a little bit at least about Eccentex. We are a low code/no code platform and actually in the past we owned and invented technology that helps the business users to create their own application by leveraging different visual builders and designers which makes the development of any application very easy. But at the recent months we just aimed to move even from low code technology to the next generation of business application development which we called No Code and we took a big step forward to employ a new AI technologies that enables us to create the business application in certain cases even without having a single code without needing a single code. And we already deployed couple of Eccentex AI application that can help the business users to adapt some new functionalities which actually works based on a single click and we will see some demonstration from Max about that new features. So all applications what we are building will be equipped with the new AI technologies embedded. It helps the customers and the business users to search, analyze or even reply to certain cases and even invent a new solution by a single click.

So this low code technology is extremely important especially in the near future because as we learned from George F. Colony. from a CEO of Forrester recently presented a very interesting bold statement in the Innovation Technology Conference by saying “The most popular programming language of the future will be English.”

This is a very broad but most probably a valid statement.

And Eccentex is adopting the development and the technological road map accordingly and our aim is high, so we’d like to help the business users to apply and deploy technology and business application based on No Code Application

about 60 percent of the time and using as an additional low code platform technology to augment the solution or make the solution more tailored by 35 percent of the time and only need five percent or so just to use a deep code. Then further customization or some specific application interfaces need to be developed so that’s our aim. We know that this is very challenging, this is very high, but this is the roadmap, but we’d like to follow in the next couple of years.

And very importantly we came up with a new offer called HyperAutomation Cloud.

Based on the redefined terms of hyperautomation which was originally defined by Gartner in 2020 and was called as the number one trend in the strategy business technology at that time, and it’s still valid and our definition is close to but a little bit more than the Gartner defined at that time because it do believe that hyperautomation is definitely not just RPA and adding an AI capability to the robotic process automation features, it’s more like an orchestrated use of multiple technologies, tools and platforms together such as AI such as machine learning every and software architecture and even business process management, operational CRM,

Knowledge bases, automations in any way that’s all together defined a new way how HyperAutomation can deliver and we are serious about to execute this and our first step was couple of months ago we successfully trademarked HyperAutomation as a Service (HaaS) so we own this new business category and we would like to make the business user’s life even more easier not just simply allowing to use our platform which helps them to deploy, create, deploy and even design new business applications which is all embedded with some AI features but were actually able to operate that on behalf of them. Our aim is to deliver their customer experiences on behalf of our customers to their end users by using the technological platform we have and adding a service on top of that which including operational and maintenance and also the application development services. That’s why we invented and deployed this HyperAutomation as a Service as a concept which is available for those companies who would like to leverage the IT operation through us.

How it works together because those ones who are following us may already noticed that we launched two new platforms two new domains the Eccentex.ai and the Eccentex.io on the top of the Eccentex.com which is our traditional platform and why we did it because we are segmenting our services in three different strategic categories. So the most important is eccentex.com is responsible for all services and publications and the platform which is operates as a core of the Hyper Automation concept and the professional services part of our business is running the Integration, Deployment, Configuration services and they’re also exposing all these open AI connectivities through the eccentex.io domain and our eccentex.ai domain is collecting and offering all Cloud Ops and AI operation services which included in our portfolio and we would like to deliver all these services together in a highly orchestrated and integrated way and the HyperAutomation Cloud as a concept is contains the software the related services and also the all AI features which is attached to them.

I know what this is, maybe too busy but I would like to just highlight the platform and the capabilities, what we developed so far and as you can see on the middle still, the Eccentex AppBase platform is the core of the offer

which enables the customers to create, deploy and operate their business applications and also enables us to store those business applications we already developed and offer as a core application such as the email automation, the dynamic case management, the knowledge management help desk and and all these applications, which is horizontal by nature and enables the business users to deploy it even in vertical use cases on the left side. That’s the new innovative platform we created through eccentex.ai which is nothing else, just the federation of different third party and or inhouse AI what we developed

because we do believe that we don’t want to even compete or go against any best of breed AI services which is available on the market, but we would like to use all of them, integrate and orchestrate them together and expose it as one single service to our customers. So that’s why we created our personification of our AI. We call it Eccentex kAI. This is a kind of interface that the customers can leverage the different 3rd party AIs together with our AI capabilities in an orchestrated way, without having considerations about different background integrations which needed. Why we are doing this because a certain AIs what you can leverage from the market have different core capabilities, different strengths and we are just selecting the best of breed AI capabilities, put them together and expose it as one consolidated service towards our customers.

On the right side, eccentex.io , this is the Open API gateway which responsible to integrate our platform with every single CX ecosystem components such as contact centers, third party three M workflow, databases, knowledge bases, whatever needed to provide the CX. Why we are doing that because we do believe that the customer experiences exist on the level of that ecosystem. Nobody in the market can deliver these experiences alone, we need to work together, we need to orchestrate these capabilities. That’s why we opened up our platform and enables the third part integration. They are strongly believed that customer 360 desktop which sits in the middle of the experience which enables the agent with the the consolidated view about the customers, it’s very important, and it needs a lot of integration as the data components which mandatory to serve the customers during your journey. It lays in different systems and need to be even real time pulled together and integrated and displayed in an intuitive way

on the desktop. So without any further ado I would like to go to the last slide of mine introducing eccentex.ai which is a new platform component which enables our customers to engage, automate and empower their customers and their agents with our AI capabilities. And in the next couple of minutes we will see our demonstration from Max about how we actually deployed many of these capabilities. As the time is limited, we decided just to present not all of them but just almost half of them in this solution demonstration. But you will see some great examples how we’re applying our capabilities to enhance a real life use case and how we are able to provide the customers and agent experience in the next level by using our AI capabilities. So I would like to hand over the screen to Max to do the live demo. So Max I stop sharing my screen and you can take over. Hey Max, can you guys give me just a few minutes and just my internet just went out all of a sudden. Oh. I need a few minutes to.

Restart it, give me a few minutes. Apologies. No worries no worries. I’m just going back and speak about our AI services until Max is restoring its Internet connection. So.

The AI Services what we developed is all based on a real customer needs and real life use cases.

They didn’t develop the AI services from a central, let’s say inside, approach but we took some external AI use-cases which need and need to be deployed by customers based on their real life scenarios.

So many customers who contacted with us just struggled how to, for example, maintain a large amount of emails which all need to be answered in a hyper-personalized manner. So that’s why they developed the solutions which enables them to decrease the time what needed for a sophisticated and other create answer from eight to ten from 8 – 10 minutes to even to 40 – 45 seconds or so. So they also do the same AI capabilities, for example, to identify and mask out some PII personal identifiable information from an email which should not be displayed in front of an agent but still need to be captured and need to be processed but in a way, we’d like to allow the agents to able to automatically classify each and every incoming interaction without even following a script or further engage with the customers. We are able to get the priority based on the free text what what the case or an email contains. We are able to assign the priority. We are able to dynamically search from the various databases from different knowledge bases and assign immediately the best possible answer, the best possible.

Response to an incoming case automatically. We also very concerned about that how we can help our customers to able to assign a different business rules which sometimes could be complex and complicated to a certain case is just to make sure that nothing will be forgotten. Everything will be performed on time and every single let’s say customers will get their answer or their service on time as it was promised before. So we also figured out how we are able to automatically create cases directly from an email, which makes the agent’s life and the customers lives easier because they not need to be let’s say translated what it is in in the email, we are able to pick up the most important parameters for the email automatically and able to create a case without even a human touch, which still can be let’s say overseen by the agent and modifify as needed.

We are also able to do sentiment analysis this both on emails and any kind of written media and also the transcript if its go through on our concierge bot and so on. So I wouldn’t like to introduce more details because Max will do the demo. Max, are you ready? Yes. I No worries that technology is always doing this kind of thing, so I’m stop sharing and you can take over the screen. Yes.

Perhaps we should So everyone see my screen? Yes do.

The right one here with the email.

Okay so.

I want to kind of discuss the various concepts that we have around AI, so we’re going to refresh here for a second. So the idea behind our case management system is that everything here is created as a case. So this is an example of how a case management system would look like inside of a contact center Genesys in this case that can be pretty much any contact senor or it can completely standalone. Usually I like to start off with email cases because they provide a good introduction into National Language Processing AI and other various capabilities. So let me just take you through this little analysis here, but.

Here an email customer sent an email to a customer service help desk and it has essentially converted this email into a case and this is important because this case then lives throughout the entire customer journey life cycle. So as there is back and forth between the conversation between the customer and the agents or chat bot or whatever other channels that the customer should to interact and they’re resolving their problems, it’s all recorded in one case and this provides a lot more AI machine learning capabilities rather than just a transactional conversation. So everything is connected and you can figure out down the line how everything was resolved. So here we have an email customer sent a couple questions here so what they’re trying to do is they’re trying to get off from a Samsung to an iphone. They have a couple questions about it that are you know around here, and they sent them some other information. They want to order the phone a bunch of things right here so pretty complex email here with what’s happening. So the first thing that we do with our AI and we use a number of engines for this where pick and choose the best that’s available to us. Although our primary go-to AI engine that then most of the stuff is built with is using Azure Ai.

And for one of the reasons for that is because.

our Eccentex Cloud is primarily powered by Microsoft Azure and for various security enterprise use cases. Azure remains to be one of the best the platforms for our purposes and they provide a lot of AI capabilities with us.

And so we’re analyzing a bunch of stuff here, so we’re extracting product names. This is important for routing. We can then route this information to the right product team. We can identify things like the urgency in the language of this, so this is all done in real time, so their urgency is high because they want to get the phone by July 17, it’s already passed that, so this is clearly a problem language here we can translate to over 100 different languages and we’re not translating just sentence by sentence like a normal who will translate with work. We’re actually using a neural network through an LLM, which would also be able to translate the meaning of the system as well. So not just the guess sentence by sentence but if there’s a deeper meaning behind this, there’s nuances in the language, it’ll all translate this correctly, and we can do that a bit later here. Summary of either the question that the customer has requested or the entire conversation. So as you can see as the conversation expands.

More and more topic can be covered. For example, you may be asking originally about You know how do I do these technical things? But then you may end up asking about billing questions or you know a previous case and so all this can be summarized here, so the agent can quickly see what’s going on. Things like sentiments are clearly important, so angry, unhappy, positive. And again not just is this positive negative or neutral, but also providing percent in this case completely neutral but providing additional percents here, which can now later be used for fine-tuning AI.

And then topics and tag detection so the system can automatically generate you a list of you can. These tags are predefined so you can either predefine them yourselves or what we can do is, we can run our AI engine through all of your previous 100 of thousands of cases and provide you a list of 200-300 topics that are relevant that are common across these types of cases emails. So we’re able to actually provide you these categorizations, which then will help you even better understand what type of topics these emails are about by having to predefine it yourselves. So a lot of these utilities we provide out of the platform we have here so that you can simplify your lives. You have to spend a lot of professional services time figuring out and doing a lot of analysis. We do a lot of that for you through this AI and then come these what I call the color coded items and these are specific to your use case. So here.

We do it’s called Entity Extraction and this uses AI through an LLM to identify these items that are important to you. So for this use case, it was important to understand whether the people or contact information, addresses, organizations involved, events, products. All these things are highlighted here based on the use case. But if I were talking about a different type of use case like let’s say I’m a car dealership or warranty service, I would may want to start taking out the VIN numbers, car car models and other car features. If I was a hospital or or a healthcare company i’d be pulling out from, i’d be asking for your information like symptoms and medication. So I think this is all configurable within our product and you use this, you know a very powerful engine to do it and we’re not using again you can if you’re talking about extract my account numbers, you can say I want to account number with this type pattern like.

Otherwise you can say give me a list of all the people mentioned and it uses a large language model essentially GPT4 to then understand who they’re talking about and picking out these people. Okay.

And then a couple of other cool features around this AI things like PII detection and redaction. So this is our ability to extract things like credit card numbers, passport numbers, social security numbers. You even go further down and extract things like people’s age, their symptoms you know we can kind of go down here, but in this use case, this person has sent over a credit card number and before it actually got to the agent or the user, we were able to redact all this information and this can then be either completely reactive from all records on the system or can be simply redacted from agents. And if they want to see it or supervisor wants to see it, they would need to click on a button somewhere here and approve this and it’ll get recorded that they were seeing product information.

Going forward we also all these tags that we talked about as the case expands we’re constantly including proveving more so as the conversation expands back and forth we’re able to get more of these topics and then because of this keeps adding to it. And then the lastly from this simply from an agent view here is the Smart Knowledge Base capability. So for those of you don’t know, we have a built-in knowledge base system that is part of our product. It’s a special add-on and it’s a full-blown knowledge base system. Very powerful use for enterprise use cases a lot. And we have a smart context-centric search of everything that is know relevant to the case or not. So while it can be used like a simple Wikipedia where a user can search for content or links, that’s obvious use case but its more relevant use case is the ability to

highlight or suggest articles are actually relevant to the use case. So we take things like the topic, what the case is about, it can be the case can be information about this can be from the case of specific information. So if it’s a FRA investigation case it will be you know what the fraud is about. There is description about the fraud, whatever it is or in this email use case it will do an understanding of what the questions are about and it’ll start suggesting articles semantically. What are most relevant but not just semantically. What’s important here is that because we have a lot of other information here, we have things like the language. We have things like we know about the customer the email that we know who they are. We know they priority level, they know where we live, we know their previous product so we’re not going to be support. We’re not going to be suggesting warranty centers that are in Florida if they live in California, we know where the customer is even if they didn’t mention the case. So we’re able to conceptually see this information to the search engine and it’ll provide this information here that’s very relevant. Okay.

That’s the concept of smart replies here. So let me just you know that do here we have the ability to automatically generate a response to a customer’s question. So we have a whole template system, snippet system, all that is you know, part of any email help desk system where if actually to generate a response to the customer. So the agent can write know some instructions of what they want to tell the customer. So in this situation, you want to tell the customer essentially how to do what they want to do here and it will generate your response very quickly. So here we have here, generates you my path and essentially goes through all the questions that they’ve asked here and answer them for you. I can change this prompt up i can say you know also include this fedex number or ask for additional information. All that can be done here and it’ll automatically generate these responses. The agent can then go ahead and make some changes as necessary.

It’s also possible to configure the system so that they would that whatever the agent or custom or agent or user respond to this customer’s request, you can also go through its own AI engine. So for example, if the agent for some reason starts swearing or becomes rude or there are other issues here, this can all be detected before the agent before it’s actually being sent. Similarly, because of the neural network capabilities of translating things, it’s also possible within our product for the agent to respond the language that they prefer and then for a system to generate the still later specific response as well. So a lot of NLP type of NLU natural language understanding natural language processing type of capabilities within our product. For simple iquiries chat bots things like that.

The product also comes with its own CRM system so we have a a full-blown service centric operational CRM capability within our product. So it’s all configurable through our system integrates with other core systems. So if you want to bring in data about the customers from your other core system such ASAP or Salesforce or IBM, whatever it is, it’s all possible to do that and mash all that information up in one screen. So since the topic of this discussion is primarily around artificial intelligence, I’m going to focus on all thats for now. So we have, you know, this is a use case for let’s say Costco. You can see here and we can see things that call the previous orders here. We’re able to detect this kind of an overuse KPI but to detect what the customer may be calling about. So if the customer is able to see you the call, the agent’s able to see right away highlighted things that are probably important for the agents know before they even pick up the call. So we know that the person may be calling about an email that the customer sent a it earlier or a late delivery that has happened. So again using artificial intelligence we’re able to highlight things or

surface information that is very relevant to the customer journey here, predicting what they’re talking about ahead of time and trying to get the agents acquainted with the situation so that everything is done faster and more smoothly, and that the customer feels more personalized, that actually someone knows what the problem is.

Another part of AI right here is the ability to,

Use all sorts of information we have about the customer and all the previous inquiries or cases. Whatever happens to provide maybe offers or next best actions, and these are you in this situation because I kind of made this more into a sales use case, we have here offers. So we know what the person has purchased because we integrate with whatever core systems that the customer may have.

We also know all their previous email inquiries we know the topics they talk about know we know how the previous case will resolved. So if a customer has styled a complaint against about internet speed in a travel that they recently had, then we can make sure that we don’t offer that type of service again or whatever else we we can kind of machine learn off all of these previous interactions with the customer as long as demographic research as well. So we know I’ say this person’s a gold member. We know that they joined for this many years and their spend for this year is this which is a quarter less than a quarter of what they typically spend a year on all of this. So we’re able to then start providing, let’s say, offers that are more relevant. We can say you know that they recently bought at TV so and typically their type of person that would be willing to buy this type of expert plan protection for the TV or second membership because we know that they have a spouse that need that may also want out membership purpose like that. So we can try these personalized type of offers again based on effect that we have all this information all consolidated in one place. We can use the machine learning capabilities to predict all these things.

So how was all this built? This all kind of looks almost tailored through a specific use case and that actually is the case I mean behind all this is a very powerful low code platform for case management and for operational CRM that we built and all of this is configured relatively quickly. So go into, let’s say, the various setups of how all this is done. Then we can see that we can design, let’s say our customer. So our customer information here is all modeled very nicely similar type of case you want to do. It’s a front investigation case would have its sell data models and workflows and things like that. And so you can model this entire database or your model your customer. And this information doesn’t have to all live in our system. You can just model it as though it does. And then there’s various ways to make sure that the data is either synchronized with our system or that we call.

Core systems to get that data as needed. But you can design all this here and then because it’s all designed here. It makes AI use case machine or any cases much more easier, so we can just assume it’s all here. The AI results can easily be shown anywhere around the system, so we can have, let’s say here the Aicenttics AI widget that you saw for the offers is, you know, we drag and dropped it over here, configure somewhere around the product. We have workflows. We have all sorts of rule builders that can be that can use all these AI extraction methods. So for example, if we could give it heres a simple AI triage rule builder that we have here. So let’s say an email comes in or a case with of information comes in and we say that the language code is English. The sentiment is negative or the partner or customer is in some sort of network true. So by designing this type of condition we can say What do we want to do? We can say that we want to increase the cases priority because of this AI stuff. We want to send a special type of acknowledgement, not just a normal acknowledgement but a DIP acknowledgement here.

And I want to assign a case to VIP team. So again using things like sentiment, language codes or other types of indicators predictions about the case or the customer, we can then use and configure here very easily. Max I just would like to add very important things which need to be highlighted. So this is a great example showing that the AI analytics capabilities is directly integrated with the rule engine. So whatever the AI define find in the text or in the analytics phase, it can trigger an action on the rule engine and it it’s integrated altogether in that system. Right. Yep, absolutely.

And this is this is one of the things that know always question about AI in general is the AI boom has provide a lot of these capabilities, but there’s always a question about it. Well, how do you actually use it in the real world? And of course, you can integrate Chat GPT and have it create knowledge base articles for you. But you might as well just go into Chat GPT, write what you want to copy past into your knowledge base. This where we provide value. The value that we provide is because all this information about the customer, the cases, the context of your system, the rules that you currently have are all in one place, the stuff we generate for you, the rules we generate, the UI, the machine learning. It’s all based on this contextual data that we currently have. And so this provides actual use cases that are important through writing, just copy pasting things between different systems. So.

We can very easily kind of transition now to.

The newest coolest features we have which is around the what we call the solution inventer capability of actually using a narrative to build a system like ours. So as you saw, there’s all sorts of configurators and page builders and workflow builders throughout our system. But you know this just takes a bit work to configure, so we’ve done here is we’ve created the ability for a customer to a user administrator to explain what their case management system looks like. What is it? And you can know send you right up to 20 pages of text here. Copy paste an RFP whatever you want. But you can describe it here and you can see

the results. So here I say I want to create a complaints case managed system for coding retailer and it will generate you a number of case types. So I just asked it to generate one for now but says generate you have product quality complaint case type, the workflow that’s in our system. Explanation what everything is the various resolution codes, the form fields which would then create the data model, the forms that are necessary to initiate this type of case type, the people involved. So the complaint and the case manager, the product quality investigator. This will create all sorts of operational CRM data models and forms and search screens or reports and all that for you.

And then the various reports that may be involved in all this thing, so create you the complaint summary list of all, the

you know complaints that’ve been happening the product, the product categories that they’re related to, the.

With customers maybe’re having the highest amountous complaints or what categories they’re in so you can of create these reports right here based on these types of case types as long as a bunch of other things through our systems like the email templates the

may be involved in this so these are automatically then attached to the work flows here so generate let’s say a complaint received type of letter template. Well automatically attach that event to the milestone here so that when a case is created that letter template would be sent. So this gets you pretty far ahead so this doesn’t generate you an application you can just go put into production usually not unless in a very simple business, but it will get you there very closely. It will provide you a good starting point to provide all the interfaces for you all data models everything you may need at least get started and start prototyping. And you can start playing around. You can start explaining your use case more explaining differently , have to change some things around so that actually fits your use case. And once that’s done you can use there’s I hit the button here in the demo because people keep pressing and I like the case types, but in general I have a button here says Create entities and this will actually create all this in our system. And then you can go through the various builders here and start adjusting them as needed.

So a lot of AI stuff. We have a lot of things on our roadmap as well.

Around advanced Document Processing.

Around understanding more of the language that is occurring. Labeling, texts and other machine learning capabilities are just pretty much going to be the core focus of our product going forward. Everything is going to be AI enabled and we’re eager to work with many customers on specific use cases you may have. So what we’ve done is we’ve done a lot of work about integrating these products, these AI capabilities, our product, but we kept it flexible. So if you have a specific use case that you may want feel free to reach out more like we’re able to configure it to do all of that for you. And then’s five minutes left, I’m going to hand it over to Tibor. Thank you Max, absolutely, it’s gorgeous, so I’m always learning even if you I already seen this presentation couple of times. But this is really great because because this new AI features is adding to our system almost in a a weekly basis. Do you see my screen? Yeah.

Oka so this is the way how you can engage us and if you have any question please feel free to send it either to our email or just visit our website and send your questions through that or you can directly call our sales team. You can watch demo. You can follow us on LinkedIn even even ask a question through LinkedIn or directly from me or Max and yeah please visit our website and try to try to think about that. How this use case is what Max is just showed how these capabilities can be applied in your business and even if you start small just enable these capabilities and start automating something.

Which is process related. You can see enormous amount of savings in time and effort within a very

short time without doing any activity which risking the operation of the system or just needs a massive IT investment. So these use cases are able to be added to your existing system. So do not worry if you have some legacy systems or if you have any kind of all the application which must remain in the service, we are more than happy to integrate with all of them. We are not encourage you to replace your CRM, BPM or any other solutions because we do believe that the orchestration and the integration of the existing platform is also important to provide the customers

and the agent experiences together. What we are offering is more like a help to move either from your premise solutions to the cloud or reintegrate those solutions in the cloud, or adding new features to your system. So many customers of ours are selecting some business critical use cases, apply our system, then when the system is proved itself, they are growing by adding new processes and applications as they go. So this is not a Giga Mega platform you need to buy first, then apply everything in the same way. You can feel free to start with one or two use cases, then grow from there and also this new AI capabilities you see if you think about that, how it can be applied in your in your day-to-day practice can be used also on the front office and the back office side, and also on the customers and the agent side equally and one additional capabilities we are really proud of. We are able

to integrate it inside in a Genesys desktop in a contact center desktop which is kind of unique for us, because if you think about that any other CRM and BPM and Workflow vendors, they are always wanting to keep their own desktop, and they are not really allow you to put anything inside in their own thing in the other third part the omni channel desktop, but for us, it’s not an issue. We can place our solution inside in the dynamic iframe, which means that you are able to use it without changing the customer or the agent desktop. Without even changing your applications, we are able to be embedded. Consider us

more like an ecosystem and orchestration play and of course we are more than happy to provide further information. So thank you for watching this webinar and I’m hope that you have some questions and we are happy to answer that. Please feel free to connect us. Watch demo or ask more questions. I will make sure that all of you will receive the recording of this webinar and also the presentation slides, and I also will attach a demo link to that with showing that how you can actually leverage these capabilities in their real life. Thank you so much. Thank you Max for presenting this. Thank you for your participation and I hope we’ll see you next time. Then we go even further in these use cases. Thank you.

Have a nice day.