In this episode of RPI Tech Connect, host Chris Arey sits down with Myles Coates, Cloud Engineer and “solutions expert” on the Yoga team at RPI Consultants. Together, they discuss the overwhelming landscape of enterprise software and explore how Yoga Flexible Software is cutting through the noise with AI-driven, tailor-fit solutions.
Myles shares real-world examples, from automating vendor alias recognition to processing over 100 million documents with custom classification models—and shows how Yoga listens, learns, and evolves with its users. You’ll also get a glimpse into how customer feedback directly shapes product features and why “flexibility” isn’t just a tagline, but a core philosophy.
Whether you’re a tech leader or just software-curious, this conversation offers a sharp look at what modern enterprise solutions should be doing with AI.
Interested in listening to this episode on another streaming platform? Check out our directories or watch the YouTube video below.
Meet Today’s Guest, Myles Coates
Myles Coates is a Cloud Engineer at RPI Consultants who specializes in Azure cloud architecture and AI model development for the Yoga application. With over 10 years of experience in technology, Myles brings deep technical expertise and a passion for understanding complex systems to deliver innovative solutions that drive efficiency and performance.
In his current role, Myles manages cloud resources, contributes to the application’s codebase, and maintains the SQL data warehouse, ensuring seamless integration and robust data management. His approach centers on digging deep into system intricacies to identify root causes and implement effective fixes, reflecting a commitment to quality and continuous improvement.
Beyond technical work, Myles actively engages with the tech community through webinars, social media, and in-person events, sharing insights and fostering collaboration. Motivated by a strong dedication to his team, he also has a personal passion for automation, working to streamline daily life through smart home technologies like Home Assistant.
Meet Your Host, Chris Arey
Chris Arey is a B2B marketing professional with nearly a decade of experience working in content creation, copywriting, SEO, website architecture, corporate branding, and social media. Beginning his career as an analyst before making a lateral move into marketing, he combines analytical thinking with creative flair—two fundamental qualities required in marketing.
With a Bachelor’s degree in English and certifications from the Digital Marketing Institute and HubSpot, Chris has spearheaded impactful content marketing initiatives, participated in corporate re-branding efforts, and collaborated with celebrity influencers. He has also worked with award-winning PR professionals to create unique, compelling campaigns that drove brand recognition and revenue growth for his previous employers.
Chris’ versatility is highlighted by his experience working across different industries, including HR, Tech, SaaS, and Consulting.
About RPI Tech Connect
RPI Tech Connect is the go-to podcast for catching up on the dynamic world of Enterprise Resource Planning (ERP). Join us as we discuss the future of ERPs, covering everything from best practices and organizational change to seamless cloud migration and optimizing applications. Plus, we’ll share predictions and insights of what to expect in the future world of ERPs.
RPI Tech Connect delivers relevant, valuable information in a digestible format. Through candid, genuine conversations and stories from the world of consulting, we aim to provide actionable steps to help you elevate your organization’s ERP. Whether you’re a seasoned professional or new to the ERP scene, our podcast ensures you’re well-equipped for success.
Tune in as we explore tips and tricks in the field of ERP consulting each week and subscribe below.
Transcript
Chris Arey
With so many software options out there, it’s easy to get overwhelmed. But here’s the truth. If your tech isn’t making your life easier, it’s probably not the right fit. Today on RPI Tech Connect, we’re diving into what truly flexible and configurable software looks like in action.
Welcome back to RPI Tech Connect. I’m your host, Chris Arey. Today we’re talking about how businesses are often faced with so many different software solutions.
They’re all designed to make work easier, but sometimes that’s not the case. So how can organizations stay agile? The answer is with flexible and tailored software solutions.
Whether you are scaling fast or navigating complex systems, having the right tech can make all the difference. Joining me today to talk about all of this is Myles Coates, a cloud engineer here at RPI Consultants. Myles, it’s great to have you on the program, sir.
Myles Coates
Hey, how are you doing, Chris? Thanks for having me. Long time listener, first time caller.
I’ve been working here at RPI for about three years. My title is Cloud Engineer, but on the Yoga team, we really like to call ourselves solutions experts.
So what that means is, regardless of your title, Cloud Engineer, senior developer, junior developer, we all identify problems and find solutions, whether that means contributing to the code base, setting up and maintaining resources in Azure, or talking to clients.
Chris Arey
I love that you made that distinction clear and upfront because that is a major theme in today’s discussion. So I love that you’ve already set us up on the right foot. Without further ado though, let’s get into it.
You mentioned something there, Yoga. It’s our proprietary software. I’d love to hear from your perspective, what does it do? Can you give us an overview of how it fits into the larger ERP ecosystem? Yeah, what can you share?
Myles Coates
Sure, yeah. So, Yoga is a cloud-based enterprise web application specifically designed to enhance enterprise systems by extending ERP capabilities.
Our application was built from the ground up within RPI. It utilizes custom code and AI technology, including tools from Microsoft and OpenAI, to streamline operations like AP processing, invoice extraction, and vendor communications.
Our software can be broken down into three categories, Yoga for FSM, Yoga for HCM, and Yoga Connect for legacy systems. The key advantage of Yoga is its configurability.
We can fine-tune Yoga to meet any unique business need, whether that be automating certain processes, creating unique rules for routing and learning, creating custom classification and extraction models, or even offering custom logos and usernames in your Yoga Enterprise instance.
When you log in, you don’t see Yoga, you see your actual business logo inside the application. It’s just the little things like that.
Chris Arey
Yeah, I love that. I remember when I interviewed at RPI, I was over two years ago now, but I heard the name Yoga Flexible Software. I was like, wow, what is that?
And I’ve heard throughout my conversations with folks that it’s this tool that fills the gap. Is that accurate to say?
Myles Coates
Yes, yeah, definitely fills that gap, that missing gap.
Chris Arey
Cool, I’d love to hear more about how it does that. But my next question for you here is, you know, we started this episode by talking about how there are millions upon millions of software solutions out there right now. No shortage of options.
But how do organizations really decide what’s going to work best for them and in what way does Yoga kind of help them address these challenges they might be facing organizationally?
Myles Coates
Yeah, yeah, kind of like I mentioned before, Yoga prides itself in being highly configurable. We take the time to learn an organization’s workflow and adapt the software to fit the unique processes.
Rather than forcing them to throw what they know out the window and learn a new foreign predefined process with this new system, we really pride ourselves in trying to bridge that gap and make it as easy as possible to enable that in transition.
The software is designed to integrate smoothly with existing ERP systems, offering customizable AI-driven workflows that learn and adapt to users’ habits over time.
Chris Arey
I love that, Myles. Would you by chance be able to share an example with me?
Myles Coates
Yeah, sure, sure. So let me give you an example. So let’s say I am a user that’s processing documents within Yoga. This can consist of uploading like receipts, invoices, packing slips, any type of like those type of documents.
Our system OCRs those documents and is able to extract key information. We’re talking about vendor names, line items for packing slips, PO numbers, finance totals, invoice amounts, and things like that.
So as the user, while I’m spot-checking the documents to ensure that they are accurate, I noticed that I need to change a specific vendor name. Let’s call them vendor A. Now the Yoga system, it correctly OCRd the vendor and extracted the correct information, including the name, which is vendor A. But I know as a user that for our business, we reference vendor A by an alias in our system.
This is where some of the true flexibility of the Yoga application comes into play. So Yoga, it will automatically identify that for the last three to five documents containing vendor A, I have had to manually change the vendor name field from vendor A to its alias inside of our system.
Using AI, our application will automatically generate a custom rule based on a parameterized prompt to ensure consistency for this change and will suggest that rule to me, the user, on the name field whenever it detects that I am in a document that is from vendor A.
Then, as the user, I can review this rule that was generated from Yoga, which is presented to me, and I have the choice to automatically apply this rule going forward to all documents coming from Vendor A.
So I will no longer have to go in and manually change that. Yoga said, okay, I’ve seen him changing this over the past couple of documents. I’m going to create my own custom rule and I’m going to present it to the user to make it easier for them.
Chris Arey
You know, that sounds really cool. And like over time, that’s going to save people a lot of effort from having to go and manually change these things each time.
When it makes that suggestion for the first time, if you accept it, does it give you the ability to accept it for all future invoices after that point too? Wow.
Myles Coates
Yes, And yeah, as you said, it saves a bunch of time because this rule is made in real time by the Yoga system. As an end user, there’s no need for me to open up a ticket and explain the business logic and other nuances as to why I need this rule set up like this.
Maybe wait a couple of hours to a day for a dev to create and test that rule for me. Then I have to wait for that rule to be moved into the application after it’s been tested.
So we’re really saving a lot of time here, making the process more efficient, and putting that extra time back into the end user’s day. That’s just a small example. Yoga can handle varying levels of complexity in AP processing, vendor management, and HR data. I mean, the possibilities are endless.
Chris Arey
That’s really great, Myles. Just like hearing how the application studies user behavior to make these smart recommendations, like, that’s like cutting-edge stuff, right?
Myles Coates
Yes, yeah, yeah, exactly. I mean, it puts the power in the time right back into the user’s hands. It’s like having an assistant kind of with you at all times as you’re working through documents and processing.
Chris Arey
Taking notes and saying, you know what, you know what would make this better? And then it suggests something.
Myles Coates
Yes, exactly. Exactly, yeah. It’s like, you know, Myles, I see you’re wasting all this time changing these fields for every single document. Why don’t you just do this? Sure, of course. Thank you.
Chris Arey
That’s great. So I’m hearing a little bit about AI and the role that it’s playing in this situation. Obviously, AI is all the rage. That’s no secret. It’s everywhere, always. Maybe more than people would want to see, but that’s beside the point. I’d love to hear more about its role in shaping the Yoga product. What can you share?
Myles Coates
Sure, yeah. So we use a full suite of Azure AI services, including Azure AI Search, Azure’s OpenAI models that are given to Azure from OpenAI such as ChatGPT, the Azure Document Intelligence Library, and pre-built and custom that we’ve trained ourselves, extraction and classification AI models.
Chris Arey
Okay, so some pretty extensive AI usage then it sounds like.
Myles Coates
Yes.
Chris Arey
Any stories you can share about how AI has made a big impact for a customer?
Myles Coates
Sure, yeah, yeah. I can talk specifically about one of our custom classification solutions that we built. Basically, our client had tons of PDFs. I’m talking like 100 million, 100 million plus PDFs, ranging in pages from one to 1000 plus pages.
If you think of, yeah, just a few, just a few. And they needed to get all of these 100 million plus PDF documents ingested into OnBase. They needed to extract key information such as employee name, employee ID, custom document types, document dates, etc. from these PDFs to properly sort them and store them inside of OnBase.
So we can imagine if they had a person or a team doing this. Chris Arey (10:42.73) full-time job for a long time.
Myles Coates
Yes, or they’re talking months to maybe even years of just reading through PDFs, typing down information, and sorting them into different folders.
Chris Arey
Is that what they were doing?
Myles Coates
That’s what they had to do. Yes, yeah. So you can see why this is a helpful solution. They were very happy with the solution we came up with. So basically to build out the solution, they provided us with a list of document types and a few hundred examples for each.
That’s it, just a few hundred. We took those documents and trained a custom classification model to assign the correct doc type.
Chris Arey
Okay, I’m listening.
Myles Coates
During our testing and design phase, we trained our classification model on batches of a thousand documents at a time provided by the client until our overall classification rate was above 75%.
Chris Arey
Classification rate, is that like success?
Myles Coates
Yes. So basically what that means is they were their baseline that they wanted was that for every 100 documents scanned at least 75 of them are 100% accurate.
Chris Arey
Okay, got it. So that was a threshold that they had internally established. And so you built this testing solution to meet that.
Myles Coates
Yes, I mean, I believe with a few more weeks of testing, we could have easily had it classified at above 90 percent. But the client was happy with seventy-five. So that’s what we provided.
Chris Arey
Seize grades or… Okay, nice. So then as a result, what happened? I’m hearing in real time that they went from this very manual, probably exhausting job for a team to something that’s happened instantly. What’s the final product?
Myles Coates
Yes, so like for the finished polished product, we used our application code to split those large packets of PDFs into individual packets split by doc type. And we did this by utilizing the Azure pre-built read model to extract the key information such as employee name, ID, and document date in combination with our custom classification model.
That is to assign the correct custom doc type, and then we also have our custom coding and logic to merge those two together. So what the client gets is a finished product and export of document-type folders with the PDFs already pre-categorized.
So literally, all you have to do is take that exported folder of our files and copy them in their own place. That’s it, nothing else.
Chris Arey
Into those folders. Wow. Man, so you saved them maybe an hour or two. I love to hear that Myles, and I like the way that AI was being used in that situation.
And so I have to ask. You know, this was a unique request from a customer, they needed to solve for this problem. And you listened to the feedback, you took action, and then you built that right into the Yoga solution.
Would you say that that is a common outcome for the Yoga team is that you receive feedback, you build it into the functionality standard, and then now other businesses can use that functionality too?
Myles Coates
Yes, that’s actually a key part of our development. We’re always listening to our customers. We’re always trying to hear what they need. And oftentimes, we hear common themes amongst different clients.
So in the past couple of months, we’ve been leaning into how we can take the AI that we’ve been developing and use that to make the user experience and the Yoga software itself better for all clients.
We’ve developed and rolled out a lot of exciting features that have mostly been from client feature requests.
So these aren’t things that like we found and we said, hey, it could use this, hey, it could use this. No, these are things that clients said, hey, like this would really make my life easier. And we took that and we say, you know what? That would probably make everybody’s lives easier.
So some of the things that we’ve come up with are the automatic creation of custom logic rules based on user interactions like we kind of talked about earlier.
We created custom extraction models to extract information from packing slips, including line items. The line items were a big ask from multiple clients. They said, hey, we have packing slips. We have receipts. We can OCR them and get the surface level of information, but we want detailed information on the line items.
So we were able to cook up that custom extraction model to get that information for them. We built classification models to detect blank pages. Now this one was not a feature request from a client, but it is something that we noticed that would improve the client experience.
So basically, let’s say we get a large packet of 30 pages. Out of those 30 pages, 10 of those pages are either blank or have just one line of data, making those pages useless.
Other companies might charge you for that entire 30 pages. We said, hey, we’ll build a custom classification model. Whenever it detects a blank page, we’re just going to throw it out.
Chris Arey
Wow. And that’s customer service right there, looking for ways to help them save. That’s really thoughtful. And also it demonstrates that not only are we working to serve customers based on their own requests, but we’re actively looking at ways to improve it on our own.
So you got both sides of the equation here, delivering a world-class solution.
Myles Coates
Yes, yes.
Exactly. How often do you get that where you’re paying for software and they offer you something to save you money from them?
Chris Arey
That’s the real deal right there. For those of you listening in, you know, we started this discussion today by saying there are a lot of software options out there. You want a software solution like this one.
One that demonstrates that people know the team is listening to your feedback and taking action on it. Fantastic, Myles.
Myles Coates
Thank you. Thank you.
Chris Arey
So AI is obviously not a buzzword in this case. It’s making a real difference here. I’d love to hear from your perspective where you see the future of AI in enterprise solutions like Yoga. What do you predict? What do you hope to see?
Myles Coates
Sure. Yes, so I’m going to say like, it’s not a buzzword at all to me. I think of AI as an invaluable tool that everyone should keep in their tool bag. However, AI is not going to flat out replace any tools in that bag yet.
And I’m saying yet because eventually I’m sure somebody will have something. Every hammer, every screwdriver, every tape measure in your bag of tools, you have an expert right beside you telling you all the ways those tools can be used.
Like, hey, you can use the hammer if you hit it at this angle right here, that’ll be 10 times more efficient. And not only is that expert telling you how the tools can be used in their best possible way, that expert is also at the same time offering suggestions on how you can use those tools that you already have just more efficiently to complete your goal.
That’s what I see AI as, is an assistant. And we apply this mindset to the Yoga application. It just opens the door to tons of future feature development. With the assistance of AI, we can find existing or potential bugs in our code faster and fix them, of course.
We can get a lot more creative on the features that we do suggest and build for Yoga. And we can build those features faster and more efficiently.
Chris Arey
I love that, Myles.
You mentioned something there too. Businesses will use assistant-related verbiage to describe their AI application, which is exactly what it is, right? It’s allowing humans to do their jobs.
Myles Coates
Yes, yes.
Chris Arey
And I had a similar discussion with Melissa Olson recently about how like, you know, people are sometimes scared that AI is going to replace them.
But that’s not really the case as long as you’re leaning into this, like now widely available resource. Like it’s a great tool. And I love to hear how you’re thinking about making better use of it going forward. So thank you.
It’s been a great discussion. Before we end though, I always like to ask my guests if they could share one actionable takeaway for today’s audience. So I would love to hear, you know, your parting wisdom here. What can you share?
Myles Coates
Sure, I’ll just talk about Yoga. That’s what I love to talk about. What makes the Yoga product so successful is that we take the time to listen and adapt to an individual client’s needs.
We are not building traditional one-stop shop software. We won’t tell you to forget what you currently know, throw it out the window, and learn our system to take full advantage of it.
And we’re not going to say, if you have a request, we’re not going to say, we understand you have this specific feature that you need or want to make the app work better for you, but we’re just not going to develop on it because that feature is not needed by all our current clients.
So who else is asking for it? Yoga remains flexible for its current and future customers. So I’m just going to leave you with this.
If you are using rigid software, that’s not customizable, doesn’t fit seamlessly into your existing workflows and is not constantly learning and growing with you to ensure that you are always working as efficiently as possible, I think it’s time to give Yoga flexible software a try.
Chris Arey
Yo, let’s go. Hear that folks? Now’s the time to call. Thank you, Myles, I appreciate that. That was lovely to hear and you’re right. If it’s not working, don’t force it. There’s probably an easier and better way.
And there are people out there on the RPI and Yoga Flexible Software teams that are here to help you continue doing the work the way that you do it.
Myles Coates
Exactly. Do it better.
Chris Arey
Yeah. Okay. So thank you for that. I have one last question and it is an important one. I don’t know if you’re prepared for this, but as a software guy, there may be a little more weight to your response here, given that background.
Is it Mac or is it PC? Who’s taking the cake? You know?
Myles Coates
Ooh, that’s a good one, that’s a good one. I’m probably gonna get grilled a little bit about this.
Chris Arey
As a software guy, listening in, Myles Coates is a software developer, a cloud engineer. So I’m sorry. What’s it going to be?
Myles Coates
Solutions expert, solutions expert. Right now, I’m really leaning towards Mac. Historically I’ve been a Windows guy. I had used Windows my entire life up until the last holiday season, but my fiance actually got me a new Macbook.
Chris Arey
So we have her to blame for this then.
Myles Coates
Shout out to her she’s awesome.
But yes since I started using the Mac a couple months ago I’ve slowly transitioned all of my workflows and stuff over to the Mac. I mean, it just, it just runs and runs and runs.
It doesn’t give me any issues.
Chris Arey
Okay. That’s fair. That’s fair feedback. Error messages and windows are, like a dime a dozen.
Myles Coates
They are, there are plenty of error messages to go around and they pop up at the most inopportune times. Haven’t experienced that with the Mac yet.
Chris Arey
Knock on wood. You’re six months in now, is that right? Not even?
I’ve had one MacBook in my life and when I first got it, had similar, you know, feelings about it as you do. But as time went on, I felt, and maybe this is my inexperience with the operating system, but I felt that the Mac didn’t want you to tinker.
It didn’t want you getting into places where you could very easily get into with Windows.
Myles Coates
I agree, I agree. But there’s a caveat to that. Like, if you start going into your Windows settings and editing the registry and stuff like that, you won’t have a working laptop anymore.
Chris Arey
Your computer can be touched and tinkered with at your own risk, it sounds like. Okay. Well, that’s great to hear, Myles. I appreciate you sharing a little bit about yourself and what computer system you like to use.
Myles Coates
Yes, anytime.
Chris Arey
Thank you again for hanging out with me today. This has been a really fun discussion.
And for those of you tuning in, if you have any questions about Yoga Flexible Software, or you want to learn more about what RPI Consultants can do for you, please reach out to us. You can do so by emailing us at podcast@rpic.com.
Again, that’s podcast@rpic.com. This is RPI Tech Connect, and we’ll see you next time. See you, Myles.
Myles Coates
Alright, bye, thanks for having me, Chris.
