Bamelak and Vlodymyr's TSFWaves Connect Antennas to Real Network Performance

Your wireless device can pass every isolated RF check and still disappoint in the real world. That’s the uncomfortable truth behind crowded stadium Wi-Fi, high-speed mobility, and the next wave of machine-type communication, and it’s exactly why I sat down with the team behind TSF Waves to unpack what “system-level wireless design” actually means.

We get into the hard split that’s held the industry back for years: RF and antenna engineers work inside electromagnetic theory and tools like ANSYS HFSS, while signal processing engineers live in algorithms, scheduling, decoding, and 3GPP-style resource allocation. For 5G and emerging 6G, especially at millimeter wave with large antenna arrays, those worlds collide. TSF Waves explains how they couple physics-based electromagnetic simulation inside the ANSYS Electronics Desktop ecosystem with a signal processing layer to produce system-level KPIs like channel capacity, block error rate, and usable spatial streams, so teams can evaluate hardware choices against real network performance.

We also talk about why edge AI, IoT, robotics, and V2X vehicle-to-everything connectivity are forcing order-of-magnitude jumps in data rates while power consumption stays a constraint. Then we explore their practical answer: a workflow-driven Python API and a “Wireless AI” agent that helps engineers run complex HFSS-based workflows without living in tedious code, while still understanding what they’re doing.

If you care about 5G, 6G, RF simulation, digital twins, and making wireless design faster and more reliable, subscribe, share this with an engineer on your team, and leave a review with the biggest wireless problem you want solved next.

Transcript

Welcome And The Problem To Solve

Roopinder 0:07

Hello and welcome to FoDES, the Future of Design and Engineering Software Podcast. My name is Roopinder Tara. On the show, we will have guests that will discuss tools and technology that engineers will find interesting and useful. Welcome, guys, to let you know. You're talking to a mechanical engineer. So a lot of things that you say about RF and 5G or 6G may go over my head, but try to dumb it down a little bit, okay? Of course, yeah. All right. So let's start out with saying me asking you You formed a company called TSF Waves. Yeah, what problem are you trying to solve, first of all? We have to ask that. All engineers are trying to solve problems, right? And I also want to know why you'd elect to do it now, because you're still trying to go for a PhD, correct?

Bamelak 0:56

Correct, yeah.

Roopinder 0:57

Okay. All right. Both of you.

Bamelak 0:58

Yeah, we're both at the University of Manitoba right now, correct.

Roopinder 1:01

Okay. All right. All right. Go ahead. Take those questions in order, please.

Bamelak 1:05

Sure. Yeah. So I guess maybe the best way to describe it is we're trying to help wireless engineers just help do their hardware designs a little bit better. So obviously, like in a cell phone, there's antennas, and that's the main way you communicate with other wireless devices. And within starting from the antenna, there's a bunch of different units of that entire radio frequency chain. And they're all involved in the design of like that entire unit, right? So I we have a specific angle that we feel like is useful. And that's based off the things that we've learned in both our masters and some of the things we're doing in our PhD right now. And so we think by combining that with the tools in ANSYS, we can help engineers better do their designs much better than previously done before.

Roopinder 1:50

Yeah. I'm a big fan of ANSYS. I've been using it from the well, not quite the beginning, but back way back when. The problem is more or less like the components of these devices don't act in concert. And that's something they try to do something with. Make everything work as a system, as it were, to optimize for now for 5G and next for 6G.

Why RF And Algorithms Must Merge

Vlodymyr 2:12

So I can take this one. So basically in wireless communication, specifically in 5G, there is two main components to it. First, there is a right, so that includes antennas, filters, and all of the electronic components you have. And the second part to it is signal processing, right? So you have on the system level, on the physical layer, there's a lot of signal processing that's going on, right? So what you have to do is that there is multiple users trying to connect to the cell tower. And what you have to do is you have to assign frequencies to those users, you have to separate them, you have to decode their data, you have to do spatial and you have to do time-domain signal processing. And conventionally what happens is that engineers that do antenna design and engineers that do signal processing design, they're different types of engineers. So if you look at somebody who works on physical layer or medium access channel layer algorithms, they have no idea about electromagnetic theory, they have no idea how antennas work, they don't know how to solve differential equations, which come into the picture. In our case, what has happened is that we basically started our research with signal processing engineers and we did a lot of algorithms. We published many papers and did some patterns on medium access channel, medium access MAC layer algorithms like the and then later on we actually moved towards more towards electromagnetic theory and actually learned differential equations and learned how antennas are important. And in actually the next generation of wireless systems like 5G, those two things have to be combined together. So you cannot anymore separately actually do the signal crossing design and do the antenna design. Now those two things have to come together, and that's specifically because for 5G, you're using large antenna arrays, you're using higher frequencies like millimeter wave frequencies. And basically, yeah, that that was our goal to try to take it together. And ANSYS has been a platform which really actually helped us a lot to really get an idea how we can actually take it into the industry, right? Because what you would actually need is that you have to implement those signal processing algorithms in order to test, okay, how much data rate can you actually transfer over this channel. But to do that in a physically consistent way, you need to couple it somehow, those signal processing algorithms with electromagnetic solvers. And those electromagnetic solvers, which do solve the differential equations, were provided by ANSES. So what we do is that we take the ANSES tools, which do the electromagnetic theory, which is solving the differential equations, and we couple them together with the signal processing that is done at the base station in order to get the data rate. And they actually take the digital twin and actually put the whole system together.

Where ANSYS Stops And We Start

Roopinder 4:42

Okay. All right. So ANSYS does both signal processing and antenna design, correct?

Vlodymyr 4:49

No. Actually, that's actually that's the point. So the ANSYS does the antenna design part, so the signal processing part is missing, and that's actually what we're bringing in. We do the signal processing and we combine it with ANSYS, and we actually develop a uh add-on tool, which is a Python interface, to the ANSYS tool, which actually uh combines the two worlds and actually allows both the antenna designers to look at the system level simulation parameters as well as the system level designers to actually include antennas into their simulations.

Roopinder 5:25

Is there a digital system processing simulation program that is like ANSYS level?

Bamelak 5:32

Oh that does signal processing algorithms you're saying?

Roopinder 5:35

Yeah, is there yeah, is there a tool like that has that uh reputation of ANSYS, but for the signal processing?

Bamelak 5:42

I think they are building some some tools, but they only have some very minor features implemented. So the reason why this problem is somewhat difficult is that as Vlad was mentioning, that the two fields were developed somewhat independently. So the skills that you need to have to develop good algorithms and stuff is very separate from the skills you need to build antennas or design hardware or simulate hardware and stuff like that. So this dual combination is it's a little bit trickier. So they do have some tools, but they just have some minor algorithms implemented, very low-level algorithms.

Roopinder 6:16

Uh that's why you got to create your own AI basically.

Bamelak 6:19

Yeah, we what we're doing is something much more complete. So we have this is things that we've learned in our master's and PhD. We we've been able to learn both sides fairly well that we can build complete systems. Some of those things that Vlad was mentioning with all those frequency allocations and stuff, it's things that are done in a network in those standards set by 3GPP and 5G and stuff. And so you're able to take the hardware that you simulate using the ANSYS tools, and you're able to test them in an overall system, which we've developed using TSF waves through Python.

Roopinder 6:50

Got it. Vladimir, you were gonna say something?

Vlodymyr 6:52

Yeah, so I thought you asked a question regarding there is other companies that try to do this, and there are. I think Nvidia has been quite successful, and they've developed this digital twin called Siona, and it's been it's actually quite widely used, and it's pretty good, but it does not have electromagnetic solvers. So it still relies on you to do some processing in ANSYS or CST or any other EM tool, and then extract the antenna fire field patterns and the other parameters of the RF chain, and then actually move them to the system level simulator. There is other companies that are pretty good, like Ramcom is doing a lot of the work that Qualcomm is right now is working on digital twin. But again, our goal was actually to really take the end because ANSYS ecosystem is the most complete in the industry right now. So ANSYS can actually do antenna design with FEM solvers, they can do high-frequency solvers like the ray tracing, they can do circuit solvers. Actually, you can basically cover uh the entire uh physics of the wireless communications with ANSYS ecosystem. And adding the signal processing layer, actually, and having flexible post-processing in Python with the interface that is easy for engineers to use, actually is we think is bringing a lot of value to the table.

Roopinder 8:03

Okay.

Bamelak 8:03

And that's a big part. Yeah, what he's mentioning is this this they have a specific thing called PyADT where you can export and do some of your hardware designs using that. And there is where you can add the algorithms that we've been able to build and everything, where you can do this post-processing.

Roopinder 8:18

Got it, got it. Okay, that helps out helps me quite a bit. All right,

What Better Wireless Changes In Practice

Roopinder 8:23

tell me this. Okay, here's my phone. I don't know if you can see it. I gotta bring it up next to my face, I think. Yeah, okay, there it is. I have an iPhone. Okay, so every few months or a year or something, I get all these updates, and most of them are not related to Wi-Fi or their antennas, right? So help me understand what is going to improve with the type of work you guys are doing on my future iPhone. How is this gonna help me? Is it gonna lower power consumption? Is it gonna make the communication faster? How's this? How's your work going to help?

Bamelak 8:58

Yeah, so there's quite a few things. Like obviously, the new antenna designs would come out with the new iPhones that come out each year, but there's a lot of things that you test for, right? And so different companies would have their own antenna designs, or maybe they have third-party antennas that they have, but there's specific things that maybe from the wireless point of view that you would want to test for. There's different ways of testing it and testing your antenna design doing that. And what we're providing what we think is a superior method to a lot of the tools that are out there.

Roopinder 9:35

Okay. Is there hope of getting the common person to understand? Okay, your Wi-Fi is going to get better with when they think, okay, it's kind of good enough. I don't really have an issue with it.

Vlodymyr 9:46

Actually, there is actually an interesting thing, is that moving into a new generation of wireless systems, actually, now with all of the AI and edge AI and all the applications and IoT and automotive and robotics and everything, now the next generation of wireless systems are really gonna be more focused on robotics and then the actually human type communications, and wireless we call it machine type communications, and that really does require you to have much larger data rates, right? So if you're, for example, your base station is to serve a factory, and the factory has let's say 100 robots and they're riding around and delivering packages for Amazon and like all the things that are or you have all these vehicles on the road and they need connectivity, like oh yeah, vehicle to vehicle communication.

Roopinder 10:31

That's been a I've had my eye on that. That sort of thing would really help if the systems were improved, right?

Vlodymyr 10:37

The vehicle to everything, vehicle to vehicle. If you look at, for example, Qualcomm, it's a good example because it's the largest player in the industry, and they have been pioneers for every generation of wireless systems. And you can see automotive and vehicular communication becomes actually a big thing because you need connectivity for AI in the vehicle, right? You have to have this vehicle connected to some edge data center, and then you want to do all the processing. And if you want to do joint communication and sensing, and if you want to, for example, have installed radars and for self-driving cars in the future, and right now for DAS systems, and uh that that becomes very important. That really requires the data rates that are order of magnitude higher than what we have right now. Okay. If you, for example, are talking about gigabits, now you need maybe terabits to really be able to support a lot of these applications, and and actually we're moving towards that. And what's required is large antenna arrays. What's required is to have not just one, two antennas, it's required to have huge frequency bands. With in each frequency band that you're going to be using, you're going to be using multiple of antennas, like maybe dozens of antennas, and all of that requires a lot of design and a lot of signal processing and a lot of simulation.

Bamelak 11:44

And even in like maybe current systems, like you can think of yourself if you've ever gone to a concert, how slow you're like maybe on a normal day, if you're just driving through the city, it's fine. But like at a concert, when they when you have a lot of users, it's almost impossible to send like messages out. So there's now it's not only an active area of research in like in the research community, but like also companies are trying to figure out ways to improve their design so that they can meet the data needs in those kinds of i okay, okay.

Roopinder 12:10

I'm starting to get it. Cool. Thanks for explaining it. Now, yeah, I'm oftentimes in a place, not so much music concerts anymore, but conferences. I was at the sphere, the sphere in Las Vegas, yeah, and we had offered Wi-Fi to everyone, but nothing was going through. It was like 10,000 people in that thing, and and I don't think anybody free Wi-Fi, but what good is it if you can't you can't use it? So that that's a real case. So it may not help me, it may not help the individual with their phones, okay, big deal, but it helps on a scale event like that, or it happens, helps industry because the machines are gonna go, machine-to-machine communication.

Vlodymyr 12:48

But also, I can give you like an example, which like in the like, for example, in the age of AI, if your phone is actually gonna do a lot of AI, and there is a lot of data that has to be transferred between the nearby edge data center that's actually gonna do the signal processing for this AI, yeah, then the data rates can actually multiply magnitude. Like even right now, the new chips that Qualcomm is working on, the data rates have to really go up because of the new edge AI application. So looking at the cellular right now, if you were to just send a message to your friend or you want to watch a YouTube video, maybe that's fine. But moving into a year or two from now, when we actually have all of this AI that's gonna be actually done at the edge, not at the data center somewhere far away, yeah. The data rate again is actually has to be much larger. So actually, there is a lot of demand for a lot of data rate. And as you said, there is power consumption issues that are also going to be coming up. And so all the things actually require you to optimize both on the signal processing side and the antenna design side, the hardware design side. And that's actually where we come in with the simulation tools that really help engineers okay to both actually get the rate up while actually using least battery power as possible. Or that's yeah, that's basically the goal. Okay.

Bamelak 13:58

And also, sorry, that's one last thing to add to that. It's also what you have to realize is the network settings that you distribution of users in somewhere like a stadium or a concert is much different than something that's happening in the city. So that's what we also try to provide, like different network settings where you can test your not only like in isolation and just test like traditional metrics, but you can test it under those types of networks, which are much different from one another.

Roopinder 14:22

Okay. All right. So you figured out that I imagine this is all an AI agent that you've created that helps with this uh.

Bamelak 14:32

Yeah, we have we're working. So the Gustavo who introduced us to you, we we are working we have a tool now out called Wireless AI, where you can do this using an AI agent. So, like a chatbot right beside you where you can set up your HFSS simulation and then run the workflows that we've

Edge AI V2X And Power Limits

Bamelak 14:48

developed at TSF Waves.

Roopinder 14:49

Uh so you figured out a way to do something better, do something more automatically using AI in the field that you're very very familiar with. Then what happened? This happened in grad school, and you said uh and you're and you said, Hey, I'm gonna make a company, and your professor, you had an advisor, baby, that said, go guys, do go do it, right? Is that what happened? You have an I think you have a professor who is in your corner, right?

Bamelak 15:15

Yeah, so we worked, we met this one professor who's an expert on some of these things. So his name's Amin Mazgani. He came from like a technical university of Munich and he it's what's hard about this field is the connection of the like the antenna space, the radio frequency space, and this information technology that we have our backgrounds in. So this combination of the two fields is like where we've gotten some of this background. And in addition to that, we've all both worked internships at ANSYS where we've developed some of the understanding on how those tools work, and that's what kind of gives us the ability to develop something like this.

Roopinder 15:48

Okay, okay. So, how did this come to be in the University of Manitoba? But we don't, I gotta I have to admit, we don't get a lot of you're the first people I've interviewed from University of Place. Do you find yourself competing with this idea with let's say that's called the big guys, like MIT or Caltech or those Stanford, right? So I think this is all or is RF or is it Winnipeg or is it Winnipeg or Manitoba? Sorry.

Vlodymyr 16:18

Winnipeg is a city, Manitoba is a state. Okay, okay.

Roopinder 16:22

Shows what I know. Show us all I know. All right. So is is Manitoba the center of the RF research?

Vlodymyr 16:30

I think there's actually two things about it which are very interesting. First, we do actually have a pretty solid antenna department here, and so the RF here is actually quite strong. And there is a quite a few professors that are well known in the field of electromagnetic simulations and the field of uh RF and antenna design. But what's more interesting is that this whole uh combination of RF and signal processing and wireless and network level design has really been pioneered in the University of Munich, and the University of Munich has been really famous for it. And there is a professor who's, I think, now retired, he's probably eight years old, who started this pretty much the whole field, and it is being a whole wave of people who've been following him. And one of his students actually just became a professor in our university, right? And and then we became his students, and then we went actually on this path, and he really led us to understand this field. And the competition here is mostly from Europe, which is most interesting because the physics background is strongest in Europe, so you can say Italy, the University of Rome. Uh, a lot of people do this in in the US specifically. There is a few groups, like one group in the University of San Diego, but mostly MIT is mostly focused on the deep mathematical side of things, information theory, that's where MIT is good at. Stanford has been very good at wireless, but there is also a lot of it has shifted to China, and then the physics side has really been Europe. So actually, if you look at the North America, the competition is actually not that great here. Really?

Roopinder 17:58

Okay.

Vlodymyr 17:59

A lot of the people who actually do what we do are in Europe, which is also interesting. And obviously, they all work for big companies which are US-based, but actually in Europe.

Roopinder 18:08

I'm an easy sell when I hear about could I could technologies and problems that can be solved. But how did you are you is the next challenge gonna be convincing VCs to fund this operation? Because they don't know how to spell RF, right? How are you gonna how are you gonna even explain to them that this is a problem that needs to be solved and you guys are the ones who are gonna be able to do it? What's the next plan here? For finance.

Bamelak 18:33

Yeah, right now, I since it's a software company outside of licenses and a few things, you don't need an extreme initial investment to get started. So it's just more so right now for us just outreaching, talking to different customers. We're we're talking to a few currently, showing some demos here and there, and just getting that started. But I think those bigger costs will come a little later down the road. So it's a we have a nice angle where we don't need a large initial investment to start.

Roopinder 18:57

Oh, that's good. That's good. Okay, all right. So you can keep developing, you can test the idea on your website. You have do you have a product that you can sell yet?

Vlodymyr 19:06

Or that you could yes, we have a product and we can sell and license. And the regarding the investments and VCs, like right now, the whole industry is really focused on AI. Everybody's talking about AI, all the money goes into AI. Again, we actually are also thinking that's important. Specifically for the automation side of things, the wireless engineers don't want to do any Python. That's just what we learned. Wireless engineers, they want to, I mean math lab maybe is maximum they would want to do, and they don't want to go into complexities of automating APIs for Francis tools. And actually, we think that this AI chatbots like that, people are developing those things are gonna be extremely important to really actually take this idea of complicated wireless design and network-level simulation and bring it to somebody who is really just a physicist or he's just an engineer who is deeply informed in ideas like information theory and is not really a programmer who can actually just sit down and actually do all of it. And I think Vamlock will show a demo, and this is something we did with Gustavo, and now we're actually collaborating with him closely to put together actually this common idea, which is actually gonna be something that is gonna be, I think, more appealing to the to venture capital firms uh in the future.

Roopinder 20:15

Okay, great, great.

Bamelak 20:16

If you still have time for a demo, I'd love to see it. Okay, should I you guys want to see it right now? It's just it's I don't know if you've seen the videos we have on YouTube. So it would just be like different clips from that to just walk you through uh how the workflow would look like.

Roopinder 20:29

Great.

Bamelak 20:30

So yeah, I'll show it. I think one thing I also want to add is it's not only that the people in

Research Roots And Startup Strategy

Bamelak 20:35

our field would they only like working on MATLAB, but it's even using HFSS and understanding what it takes to do that is also a bit tricky.

Roopinder 20:44

So yeah, HFSS. I I had somebody on the show talk about HFSS and I didn't understand a word of what he was saying. And people that talk about Python, again, I don't I honestly I think it's all this stuff is complicated for engineers. If you could supply engineers who need it with both things, hf hfs. Am I saying that right? Hfs and Python, I think you got a valuable product and service.

Bamelak 21:11

So what you have to understand is these guys are very smart and then they're very they have a very good expertise in information theory and the things that they do. But this specifically with this chatbot that we're developing with Gustavo's team, you can really export that domain of knowledge to with that bot helping you out. And so you can focus on the things you need to focus on, and at the same time have all the hardware tools at your disposal. But okay, so this is I can go through the entire video, it's only two minutes long, just kind of describing, and then I'll show you different like where our workflow comes in. Do you think this is enough time or I can just go to the specific workflow?

Roopinder 21:47

Oh no, that's fine. Let's do it, let's do it that way. Sounds good.

Bamelak 21:51

Yeah, but you can use it, right?

Roopinder 21:53

Yeah, I can see your screen. I think it's London, right? Yeah, it's London. Yeah, I have to ask you, where are you guys from? I have to ask everybody because I'm I'm not from I'm not from US, so I ask everybody.

Bamelak 22:05

I'm Ethiopian. I'm fully Ethiopian, but I was like I was raised in Winnipeg. So I came to Winnipeg when I was like six years old and just grew up there.

Roopinder 22:13

Okay. All right. Ukraine. Oh Ukraine. Okay. All right. Both interesting places. I came to America when I was seven. So okay, we have that in common. And Vladimir, what about you? You came from Ukraine probably recently when the war started, right?

Vlodymyr 22:28

No, in 2014. So it's been actually when the war started. Yeah, the first time the war started. First time the war started. Okay.

Roopinder 22:37

All right. Do you have still have relatives there?

Vlodymyr 22:40

My grandparents, yeah.

Roopinder 22:41

Grandparents.

Vlodymyr 22:42

The rest of the families here.

Speaker 22:44

The rest of the families in Canada. As wireless networks become increasingly integral to society, engineers face the challenging task of delivering high data rates with low error to a rapidly growing number of machine-type devices. To meet these demands, traditional design principles focused on isolated optimization across the wireless chain must be abandoned. Many commercial simulation tools are limited in this regard as they may not offer system-level analysis. While researchers are developing more comprehensive theories to address this issue, they often rely on simplified models for the sake of analytical tractability. Our mission at TSF Waves is to provide a holistic approach to wireless design through physics-based simulations. TSF Waves Core SM is an advanced Python API designed to integrate seamlessly with ANSYS Electronics Desktop. In doing so, it provides system-level analysis and key wireless performance indicators using ANSYS's world-class simulation tools. These tools include ANSYS HFSS for antenna design, ANSYS Circuit for designs of amplifiers, filters, and matching networks, and ANSYS SBR Plus for simulating and complex propagation environments. Using multiport communication theory, we integrate the various hardware components in the wireless chain for assessment in different wireless networks and setups. Such setups include single and multi-user networks, single and multi-antenna systems, and both orthogonal and non-orthogonal resource allocation schemes. Under these setups, we provide system-level wireless key performance indicators to assess the impact of hardware designs

Demo Workflows And Wireless AI Agent

Speaker 24:16

on the overall system. These include information theoretic metrics such as the channel capacity, block error rate, the number of useful signal streams for transmission, and many more. We offer a variety of established workflows that customers can personalize and build upon, along with many others currently in development. We collaborate closely with our customers to create tailored designs and test their devices in a wireless configuration of their choice. Join us as we advance the next generation of wireless design.

Bamelak 24:47

Yeah, so I just want to quickly just go to different parts of that video that gives an overall outline. So as you can see, so this is off of Python, and you can see that we have a bunch of different workflows. So maybe you want it like the stadium example, right? So maybe we have something for a stadium, the way like users would be distributed, the different resource allocation scheme that they'd have. And so this is where we set up like the network settings, and you can see this is like different Pi DT code to run the hardware stuff on HFSS to get the placement, which city you're working in, which kind of arena, which kind of antennas you use. And so basically what would happen is you run this code and then it would start building it. So this is a city example where we're like we have one base station antenna with two users here. And as you run press play, this thing is automatically doing it on its own and just starting to solve everything. And then after it's done, so this is just a different part of the video. After it's done, so after it's done, it will set all of it, solve it, and then we will pop out the different metrics you look for. So these metrics are based off information theory. But they've taken everything, like all the simulated antennas, all the simulated hardware into account. And then this is the kind of stuff where you can assess okay, is my hardware, is my antenna, how much data streams can I send out, how many singular values can I have, things like that, which would be relevant for your design. So as you can see, just a bunch of plots pop up. So that's more or less high level, I think some of the things we do. And now the thing with Gustavo, his team at Divergents AI, we've basically taken these workflows that which are just Python-based, and now we've added an agentic tool with it.

Roopinder 26:20

I have a question though. So are you going to assure me I don't have to learn Python? I can use the agentic.

Vlodymyr 26:26

Yeah, Gustavo's tool, I think that's gonna be easy. And then, like in this video, you're gonna see that learning Python or is not actually required anymore, which actually being something that we wanted to actually fix. And then Gustavo came along and then we're actually so happy because he saw the opportunity to actually go into the wireless market because wireless market is maybe more than half of HFSS market. And what we saw is an opportunity to really not make engineers install Python and really run anything that's gonna make their heads spin. And I just look at this video, I think it's it's really a great collaboration for us. Okay, that's roll.

Bamelak 26:59

And I've said sorry, just one thing to add to that is like I've said this before, but it's really like the people, the wireless guys that work on this stuff, they do not have like this domain knowledge of hardware design. So this agentic tool, and you you take all the work of Python and like using HFSS, and then you can export it to using this.

Roopinder 27:16

I could use natural language interface, right? English.

Bamelak 27:20

Okay.

Speaker 4 27:21

So this is yeah, this is just importing the city that you want. Uh this is all in the design that we made it from this.

Vlodymyr 27:28

No, and here's importing the antennas, basically simplifying everything to the extent that it's possible to simplify this type of problems, right? You have to still understand what you're doing, you have to understand a little bit like the at least one of the sides. I did signal processing with the antenna side, depending on what exactly type of engineer you are, but you just don't have to deal with the tedious code. I think that's a whole point of this.

Speaker 4 27:50

Oh, I think like a good enough engineer to know what you're doing. You don't have to waste time learning the tdious codes or your antennas can't play.

Roopinder 28:02

You still have to be an electrical engineer though, right? You can't have a W like we try to design antenna systems, right?

Vlodymyr 28:09

Uh not yet, at least. I think maybe we we will actually maybe achieve this at some point. The problem with like all of this is that there's just too many, too many different ways you can do things, right? Because it's design. Anything you design, it's if it's a building or it's a plane or a bridge or antenna, there's always so many different permutations of things that I think that those kind of things can never get automated 100% anymore. This wireless AI is the tool we've built with Gustavo.

Roopinder 28:33

Alright, very good. Thank you for being able to explain something that's way over my head in a way that I actually actually can understand. I can't claim to understand it yet, but I'll watch it again and I'll have a better understanding. Okay, all right. Thanks, guys. Good having you on and good luck. Good luck with it. All right. Thank you. Bye-bye. Bye-bye.

Closing And How To Reach Us

Roopinder 28:55

Thank you for listening to FoDES, the Future of Design and Engineering Software Show, brought to you by ENGtechnica. I hope you have learned of a new application or technology that will help you with your job. If you have an application you think would be of interest to other engineers, please let me know by emailing me at roopinder at engtechnica.com or message me on LinkedIn.