.st0{fill:#FFFFFF;}

David Beamonte discusses Cloud Computing and IoT 

 December 13, 2022

By  Lina Alexaki

Join Our Mailing List

We publish fresh content each week. Read how-to's on Arduino, ESP32, KiCad, Node-RED, drones and more. Listen to interviews. Learn about new tech with our comprehensive reviews. Get discount offers for our courses and books. Interact with our community.

One email per week, no spam, unsubscribe at any time.

Summary

In this episode, Dr. Peter Dalmaris talks with David Beamonte.

David Beamonte is a Telecommunications Engineer with a technical background in electronics and embedded software. After more than 20 years working as an embedded software developer and Engineering Manager for the electronics vendor TELNET Redes Inteligentes, specialising in fibre optic and IoT devices, David started working as a Product Manager for IoT and embedded products at Canonical, mainly focused on driving Ubuntu Core and EdgeX.

In August 2022, he joined Arduino as a Product Manager responsible for the Arduino Cloud.

Key Takeaways

  • David from Arduino discusses his career and role as a product manager for the Arduino Cloud.
  • The mission of the Arduino Cloud is to provide an integrated platform for developing, deploying, monitoring, and controlling connected devices.
  • The platform simplifies IoT for makers and educators, with integrations with Alexa making it more appealing to beginners.
  • Arduino Cloud allows users to remotely control and collect data from connected devices, with the addition of machine learning for complex tasks.
  • Future developments for Arduino include a focus on education, industry, and innovation, as well as improvements to the platform for better integration with the local IDE.

Resources

Watch

Transcription

Read Full Transcript

Peter Dalmaris: David, thank you for joining me in this Tech Explorations Podcast from Spain, where you are working for Arduino. How are you?

David Beamonte: Thank you very much, Peter. I'm very good. I'm really honored to be here with your podcast, so, yeah.

Peter Dalmaris: Yeah. Thank you. Well, it's morning for you, so you're pretty fresh. It's a bit late for me, but I had coffee, so I'm all good to go. I'd like to take a minute to talk a little bit about you first, and then we are going to dig into Cloud computing, IoT, artificial intelligence, automation, and lots and lots of wonderful things that we were looking a little bit earlier just before we got started.

Peter Dalmaris: So, I know that you are a telecommunications engineer. From what I've read, you are a software and hardware person, so you have, obviously, a very firm understanding in both theory and application in both worlds. You've had a long career. Now you are a product manager at Arduino, where you are responsible for the Arduino Cloud, which I am a user of, and many people probably that are listening to this podcast are users of. So, very interesting in that aspect of your work. But could you take a minute to tell us a little bit about your background and what eventually led you to become a product manager at Arduino?

David Beamonte: Yeah, sure. As you have said, I'm a telecommunications engineer. I mean, that's a very broad kind of career and studies, that very early years myself towards electronics and embedded systems and so on. That was my passion and what I liked.

David Beamonte: And shortly after I finished my studies and after a couple of transition works and jobs, I started working for a Spanish electronics manufacturer, and I worked there for more than 20 years. So, the company, we really did very, very cool things. It was not a very large company. It was very young when I joined. It was kind of a startup.

David Beamonte: We started doing many amazing things along with big companies like Telefonica, for instance, and we designed and manufactured fiber optic devices for WDM switches, GPON, and lots of cool stuff. But as an electronics company, we also started manufacturing IoT devices lots of years ago, even before IoT was a term, I would say.

David Beamonte: For more than 20 years of career in that company, as you can imagine, I had to wear many different hats and developed many different roles. So, I started as a developer, as you have said hardware and software developer mainly. At the very beginning, I started also playing with electronics and developing some hardware stuff, but I focused very shortly after in electronics and embedded software. So, even though I have developed many roles, I consider myself an engineer at heart.

David Beamonte: And during the last seven years that I was at the company, I became the engineering manager for the whole department, cross coordinating all the different activities of the design of the products.

David Beamonte: And after 20 years, I decided that I needed to do a change, and around two years ago I joined Canonical. Canonical is probably not a very well-known brand. But if I speak about Ubuntu, probably everybody knows about it. So, Canonical is the company behind Ubuntu. And there, I joined as product manager for embedded systems, mainly driving some products like Ubuntu Core which is the embedded flavor of Ubuntu [inaudible]. The tiny version of Ubuntu optimized for IoT and embedded systems. And other products like EdgeX, which is a framework to develop Edge applications, everything based on Linux applications and so on.

David Beamonte: Some time ago I decided also to do another change and go closer to the hardware, which was closer to my roots, and I moved to Arduino. I joined Arduino as a product manager for the Arduino Cloud. So, I can combine my expertise as hardware expert or hardware engineer and my hardware passion with IoT and Cloud and so on. So, that's roughly in a nutshell my background.

Peter Dalmaris: Your last 20 years in two minutes.

David Beamonte: Yeah. Yeah. It's a long time.

Peter Dalmaris: I'll divert just really quickly about Canonical because you've got a two year experience there. It's one of those, to me at least, strange companies because it's product is more known than the company itself. Like, pretty much everyone knows Ubuntu, but Canonical is kind of hidden behind it.

David Beamonte: That's basically on purpose. I mean, when I was at the company, that was something that was discussed eventually, the brand is not well-known. We go, for instance, to a trade show or something and nobody knows about Canonical. You go with the logo of Canonical and nobody knows about that. But everybody was very interested in getting the Ubuntu stickers or this kind of stuff because that was very well-known. So, yeah, that's kind of on purpose to hide the name of the company or the brand behind the real product.

Peter Dalmaris: Yeah. Of course, open source definitely or can be a very viable market option for companies.

David Beamonte: Yeah. In the past, most of the companies tried to develop their own initiatives and have everything developed in-house and so on. But everybody has just realized that community development and open development is the future. So, that integration with other platforms and having open protocols and standards is the future. You cannot be trying to find yourself alone and try to build lock-in solutions. That's nonsense nowadays.

David Beamonte: And that's why I think that the open source development has cancelled trending. I mean, it is not trendy because it's cool. It's trendy because it is profitable. And most of the companies and big companies can make business out of the fact.

Peter Dalmaris: Well, yeah, that's a very good point. I just think back, say, 20, 30 years where closed source, I wouldn't say was dominating, but it was the obvious path for technology companies. Code in particular was locked. It was like a business and industrial secret.

Peter Dalmaris: Now, companies left, right, and center are just opening up the source code and are putting it out in the community for everyone to use. There are still paths to profitability outside of that, but having your code out there is a badge of honor these days. People say, "Where's your code? I can't see your code. How can I trust you?" So, things have changed so much.

Peter Dalmaris: All right. Well, just chatting earlier before I hit the record button, we said, let's talk a little bit about Arduino Cloud, but not too much because we want to talk about Cloud and IoT technology and other good stuff in general, and get the sense of the impact that those technologies are having today in society and tomorrow, especially tomorrow, because there's so many challenges that technology today and tomorrow will be called to help us with.

Peter Dalmaris: So, let's begin with your current role as a project manager at Arduino Cloud. Could you tell us a little bit about the Arduino Cloud? And in particular, what do you think that its mission in life is? Obviously, it's a company, so we get it, there's a profitability directive there. But it's got to be more than that. Can you tell us about Arduino Cloud in that sense?

David Beamonte: Yeah, sure. Let me start with - because I think that it is interesting, I mean, just out of curiosity - you have spoken about product management and what that means, my responsibilities and role at Arduino. I think that product management is one of the most misunderstood roles around there. There are different opinions and perspectives wherever you go. The companies have different opinions.

David Beamonte: And, you know, there exists that saying that says that the PM must be the mini-CEO of their product.

Peter Dalmaris: They're always in meetings, right?

David Beamonte: Yeah. That's probably an oversimplification of what the product management role is. But it is also true that it shows some of the things that the product manager must be or must do. So, what I always like to say is that we have to own the product and be the go-to person, both internally and externally. So that when somebody from inside the company has some kind of question for another department, they can just go to me and ask me what are the next steps or whatever. So, I'm trying to be the glue between the different departments, marketing, content creation, product development, business and stakeholders, and this kind of stuff.

David Beamonte: So, the responsibility is basically to own the message about the product, what has to be said about the product. Try to own the roadmap and prioritize along with the development team, the back low hand and so on to define the roadmap and so on.

David Beamonte: And, additionally, I own the message, that's the what marketing needs help with, how we can just reach the people and the user. So, in a daily basis, basically, it's doing this kind of stuff, like attending these kind of events or presenting webinars or creating content, blog posts, web pages, and so on. But, also, just working very closely with the engineering team and work with the stakeholders.

David Beamonte: And Arduino Cloud, just getting focused on it a little bit, basically just in a sentence I would say, it's an integrated platform. It's an integrated platform to develop, deploy, monitor, and control connected devices. We basically want to make it simple for the users to create things.

David Beamonte: Basically, the first stage is development. So, that's one of the things why Arduino is very popular nowadays, I would say. I think that it has become almost a de facto standard in terms of software. Not only people are using our software for our hardware, but also for ESP boards in general. And you see lots of products that are Arduino compatible and can be developed using Arduino IDE, this kind of stuff.

David Beamonte: So, development is the first stage, and doing that in a way that is simple for the users is one of our main challenges and goals. That's why we have, of course, the very popular Arduino IDE. You can install it in your computer. We have the new version, the 2.0. We can speak about it later if you want, the 2.0 version.

David Beamonte: But we do have the online IDE that makes life much easier for developers. They don't have to install anything. They don't have to struggle with dependencies, with libraries. Everything is just a zero touch and you just go and play.

David Beamonte: And the Arduino Cloud has that way for development, not only to have the online IDE, but also you have a way to store your sketches there. For those who are not used to Arduino terms, sketches are the programs of Arduino. So, one sketch is one program for the board. So, the Arduino Cloud, basically - that's it.

Peter Dalmaris: I thought I'll show you what are we talking about.

David Beamonte: Yes. This is the latest stage. So, the first one is, as I said, development. So, this repository for sketches where you can store your sketches and you can develop, you can share with users.

David Beamonte: And then, the next step is just go directly from development to deployment to program your devices with either cable or over there. And the latest state -

Peter Dalmaris: So, these are the devices. We're just looking at my account now. So, there's your hardware. This is the hardware. So, there's a hardware. These are some non-Arduino stuff as well. So, you're saying it's compatible with several non-Arduino branded devices. The deployment is part of this, if I understand your terminology correctly. So, basically, you can make the devices -

David Beamonte: Deployment is basically how you connect to the devices. I mean, just getting into terminology, devices are just the physical electronic boards and how you program them with the sketches that you have coded with the online IDE. And then, the abstractions that you use for getting the information and the touch boards are the way to interact with them. So, the process is to create a dashboard so that you can monitor and control your devices. 

Peter Dalmaris: Yeah. So, we've laid out here those concepts in the Arduino IoT Cloud. So, people can basically use this interface to combine their hardware devices with what you call Things, which actually did take me a while to get my head around what a Thing is in essence. But I'll give you an example here.

David Beamonte: Yeah. The Device is the physical element. The Thing is the abstraction that you have and you can link different devices. So, the same kind of abstraction associated the sketch, so you can use it with different devices.

Peter Dalmaris: So, there's a sketch -

David Beamonte: Yeah.

Peter Dalmaris: An example for sketch, and these are, basically, I guess the endpoints. There's a button here that is [inaudible] component.

David Beamonte: Those are the variables. We call them variables, but at the end of the day, that's the data that you get from them and how you interact with the Device. And then, you have the Dashboards to interact with them.

Peter Dalmaris: And then, you take those inputs and outputs and you can present them to the human user via a Dashboard. And here you've got also the mobile version of the same web dashboard, so you can use mobile device to access your gadgets.

David Beamonte: You're actually an advanced user of the Arduino Cloud. That's nice to see.

Peter Dalmaris: Yeah. I spent a bit of time with it. All right.

David Beamonte: The main goal of the platform is simplicity. If you think about it, you need to be loyal to your principles and to your ideas and your vision. And Arduino is very popular for do-it-yourself and IDE accessible for everybody.

David Beamonte: So, just forgetting about the Arduino Cloud, the traditional view that users have of Arduino is that it makes it very easy and accessible for everybody. And this is why it is so popular. The Arduino Cloud basically shows that. So, we want to keep that goal, and to try to make things simple, and to try to help users along the way.

David Beamonte: Because if you think about makers, you can build things. You can't just buy the things that you want some of the times. Some of the things that you want to do, it's inevitable to do it yourself. But when you create certain things, it is because you want to do things on your own and you want to have fun doing that.

David Beamonte: So, the result is important, but the journey is even more important. So, having fun along the way and being able to create something that is cool that you can show people and you can feel very good just showing it. This is very, very important for makers, I would say. And that's one of the important things of the Arduino Cloud.

Peter Dalmaris: Yeah. There's something about saying "I made this" that I kind of find very exciting. Even if it's just a blinking thing, "I made it."

Peter Dalmaris: So, Arduino's history is rooted in education, right? So, the founders did come from an educational background and they were looking for tools and methodologies to help the students learn, basically, this, which we're looking at now. And, yeah, one thing leads to the other. And I find that Arduino, like Apple - I'm an Apple enthusiast and Apple user since I was a kid - again, its origin is in the education sector, which is a philosophy that's being carried forward and it influences this simplicity that you were talking about earlier.

Peter Dalmaris: Like in my experience as a user and as an educator, I know that, for example, this can be done without any programing. For anyone watching this that haven't noticed, perhaps, it's interesting to see that, for example, when you click on this button and you add an input/output Thing here, a variable, then the Arduino Cloud in the background will automatically add code. You can go and think with the code, of course, but you don't start from scratch, so the tool itself helps you to be a programmer.

Peter Dalmaris: And when I was adding things here and then going to the sketch to look at the changes and looking at the code that the tool added to my sketch, it actually started making me a better programmer, because now I could see how a professional would write a few lines of code. And I would actually learn interactively from that without having to look at the documentation. I'm not sure if -

David Beamonte: You're touching precisely the point. If you think about this, there are many electronic designers and hardware engineers who are scared of programing. And with Arduino, they feel that they can program anything. I have many examples in my previous work. My colleagues and so on, they were just horrible developers. And they find these kind of tools very convenient because they can create just a template for them and they just need to fill with a couple of very easy to use functions, and they can create anything with that and the other way around.

David Beamonte: So, we have such a broad catalog of solutions and ideas of people who have created things that just software developers feel very comfortable connecting hardware things. Just buying the board and buying the breadboard or something, just plugging things and connecting a couple of cables because it is so detailed everything there that they feel very comfortable. So, it helps both electronic and software engineers.

David Beamonte: And one of the typical things that I'm usually asked about this, why not zero code? We cannot say that we are zero code, so you always need some kind of programing. There are some initiatives to work, for instance, with visual tools like Scratch. There are several other ones that you can use on top of Arduino. They are nice. We, of course, encourage people to do that for education purposes and so on.

David Beamonte: But it is true that we always like keeping the essence of still having to code something, because I think that it is important for people to just interact with a couple of functions with the iOS, or with whatever, because that helps you learn something else. If not, it's like just moving boxes and so on. And I think that you lose a little bit of the contact with the reality, to be honest.

Peter Dalmaris: I agree. It's like an analogy that I use here, it's being a car driver. People get to learn how to drive a car without ever having to look inside the engine or, you know, how the steering wheel works, or what happens when you press the pedal. And I feel that everyone would be a better driver if they were able to change their engine's oil themselves.

Peter Dalmaris: And the same thing with using a computer. I find it fascinating today with the ease of computing, you've got it on your hand, like in the palm of your hand. Your computer is gooey. You don't have to do any programing. You don't really understand what's going on. You could be a much more efficient user of this amazing technology if you knew a little bit of Python, maybe a little bit of JavaScript, even understanding how, say, Zapier worked, which is zero code, but you still need programmatic and algorithmic understanding they put a solution together. So, how powerful would that be? So, I totally agree with you, you do need a bit of exposure to the code and not be afraid of it.

David Beamonte: So, our goal is to simplify that to the limit. As you have shown, to create most of the code that you need so that you don't have to repeat everything every time. And if you do that very well, your functions are automatically generated and so on, but to still allow you to add stuff.

David Beamonte: Additional things that we can find with the Arduino Cloud - because we are showing here the coding, the dashboards, and so on - is integrations. You can interact with the Cloud, not only you have a mobile application, you can interact with dashboards using your browser, but you also have an open API so that you can integrate with your platforms. And we have several out of the box integrations.

David Beamonte: Like for instance, Alexa is one of the most popular ones. You can do cool things with Alexa at home, and that's very popular.

David Beamonte: This is just a curiosity. My son is seven years old. I'm an engineer, so I want him to become an engineer. But he is not going to become an engineer, for sure, but I'm trying. So, I'm trying to show him several things about connecting hardware and switching on and off certain lights and so on, and he was getting very bored about that stuff. He's not interested at all.

David Beamonte: But when I showed him that all those things could be integrated with Alexa, then everything changed. He said, "Okay. This is cool because I can just tell Alexa to switch on and off what I have connected here." I mean, these are the kind of things that I think that are very useful for people, but also very cool for beginners.

Peter Dalmaris: Yeah. It's important what you said, you need to know your audience. Sometimes the audience is your customer, sometimes your son. You need to know what makes them click.

David Beamonte: We have a very large audience, it is also true. There are many beginners who have no idea about what Arduino is. In fact, they want to start doing cool things because they have seen a video on YouTube, or on Twitter, or Instagram, or whatever. And they have seen that somebody has created a robot using Legos and they want to do something similar, but they don't know how to get started. So, we need to have a path for them.

David Beamonte: But they are also very skilled people who have been working with Arduino for more than five years and they want to feel that things are simplified, but they can do powerful things. So, we, of course, need to have a solution for them. So, that's tricky to do that kind of balance and tradeoffs so that you can accommodate things for both kind of skillness.

Peter Dalmaris: Well, you touched on a very interesting point for me that I'd like to research a little bit more. So, obviously, personally, I use the Arduino Cloud to create educational content to teach others on how to use hardware and software, programable hardware and software on the internet or off the internet, like embedded devices. But the projects tend to be tiny compared to large scale projects, like home automation maybe would be a large one for my scale of projects. Maybe a simple watering system, stuff like that.

Peter Dalmaris: But I'm very interested to know what happens with the same technology at a larger scale or even interplanetary scale, depending on how you see it. So, maybe you're in a good position to give us some examples, not necessarily the Arduino Cloud itself, but how Cloud technologies combined with hardware embedded devices, what we call the Internet of Things, is applicable in large scale problems that will help us see how this technology scales up.

David Beamonte: Okay. So, I would probably get a step back and say, what's IoT? What's the Internet of Things? Because I think that that's a term that has been so widely used that has lost much of its meaning. I think that most of the people listen to IoT and they say - 

Peter Dalmaris: [Inaudible] the Cloud.

David Beamonte: Yeah. That's something that has been used so many times. So, for me and for many people, of course, it's just connect things to internet. So, it's like having dumb things and then you start connecting those things to the internet. Why would you like to do that?

David Beamonte: Basically, for two reasons. So, it's to get the data and to control remotely those things.

Peter Dalmaris: To monitor - sorry to interrupt you, David. So, you use IoT to monitor through sensors, collect data, somehow analyze the data, and then send instructions back to make changes on site. Right?

David Beamonte: Yeah, that's it. But the valuable thing here, imagine even if you don't send back control to the things, it's the information, the data here.

Peter Dalmaris: The collection of information on the surface.

David Beamonte: Yeah. We are shedding light on some aspects of our lives that we were blind of in the past. So, now, we have much more information about our environment, about our homes, about our industries, about everything. So, that's so valuable for everybody and for business. And so, IoT, in general, just to create those things smart from the dumb world, so from just passive electronics, is a role changer. So, this is an inevitable trend, I would say.

David Beamonte: And in terms of real use cases and so on, there are so many applications. We have been discussing about some of them before the interview. But I think that, for instance, research optimization in many of the different aspects of our lives is very important.

David Beamonte: Energy consumption at home, for instance, it is very typical that everybody who starts with IoT at home and do-it-yourself, they start controlling their heating system. That's the first thing. Or switching on and off their lights because you want to save money.

Peter Dalmaris: I'm obsessed with that application.

David Beamonte: That's a very domestic issue. So, if you expand that to industry or to environmental control or agriculture, that goes a step beyond.

Peter Dalmaris: Next level.

David Beamonte: In industry, for instance, nowadays it is very, very, very important to get information about every stage in the production chain about what's going on there so that you can have zero defect production, predictive maintenance, this kind of stuff, combined with AML and so on.

David Beamonte: But also in agriculture, we have hit the eight million population a couple of weeks ago, I heard. So, we need to feed people, just optimizing resources. So, it's very important to create agricultural systems that are based on data. And there are many companies that do that. And many countries that are, of course, also using that very extensively. Like Israel, for instance, they are the number one on that. So, yeah, there are so many applications.

Peter Dalmaris: So, you're saying this is the technology that is at the core of industries. New industries, maybe like food management, not just production, but, say, reducing waste, that could be a problem that can be solved through IoT technologies. Better transportation, preservation, and then, of course, food production. We talked about earlier, robots actually becoming farmers and improving yields in crops by just removing weeds.

David Beamonte: Yeah. Of course, supply chain and distribution is also a very important thing to optimize resources. I mean, you optimize many things like oil, these kind of things.

David Beamonte: It is coming to my mind, I think that we discussed that before the interview about one curious example of a company that I worked with in the past. They were developing this kind of vertical farming inside containers, inside cargo containers. They just reused them for bringing crops and growing a specific kind of vegetable and food in places that it is not possible because of limited resources like water or sun. So, this was a Swedish company and they started just creating these kind of containers for growing vegetables in the Nordic countries of Europe.

David Beamonte: And that is very interesting, I mean, when you are doing something like that, you need to do it in a way that you can compete with the food that is grown in a traditional way.

Peter Dalmaris: That is purchased.

David Beamonte: So, you need to optimize the water, but you need to do it in a large scale. And they were looking at every single data that they were receiving in terms of lighting, in terms of watering, in terms of ground moisture, these kind of things.

Peter Dalmaris: Is that also an application where human intelligence is starting to be replaced by artificial intelligence in the sense that humans have been farmers since they essentially became humans. I think before that, we were Neanderthals or something else. Then, agriculture created human civilizations because it allowed humans to settle. Therefore, we had farmers.

Peter Dalmaris: Now, farming knowledge passes from generation to generation. Now, it is at an industrial scale and it is deep human knowledge, combined biology, and also some other things like mathematics and even astronomy in some cases.

Peter Dalmaris: Are we now starting to see the first machine AI farmers that actually can learn, not just feed them with human farming knowledge, but that they can learn from their environment and from the data that the sensors produce?

David Beamonte: Probably, yes. I mean, that's very ambitious to say. We are very good as human. We are very good imagining things. I mean, we create things, we create [inaudible], but we have very bad controlling them in a daily basis.

Peter Dalmaris: We get tired.

David Beamonte: We need some help. Yeah. Optimizing the resources, we are very, very, very bad. We are awful. We can't see around that [inaudible].

Peter Dalmaris: We also make things expensive. You mentioned cost earlier.

David Beamonte: So, I think that putting intelligence to the things that control this kind of very basic stuff, like farming, for instance. But also industry or distribution is something that, of course, already is there but become a daily thing in the coming years. I'm convinced about that.

Peter Dalmaris: David, can I just go back to Arduino for a second? So, this is something that I wanted to ask you when I was doing a bit of research prior to our discussion here. And I wanted to drill into machine learning because I believe that recently Arduino Cloud added machine learning to its, I guess, features line up. You know, it's been a long way since the blinking LED and the Arduino Uno. Now, we can create Arduino Cloud applications that exhibit learning from the data that an Arduino device can collect.

Peter Dalmaris: So, I wanted to ask you maybe to give us some more information about what the machine learning add-on to the Arduino Cloud is meant to do. For example, am I supposed to use that in my home automation system? Or is it something that an industrial engineer might use for their production and the plants?

David Beamonte: It's both. It's actually both. And that's a very good point that you are saying. We think about AML as something for industry. But because it is a really great unknown for most of us, and I include myself. I'm not an expert on AML technologies. I understand the concepts, and so on. But I have not been working extensively with it. And I think that it is a very big unknown. So, we need ways to start working with them very easily in our daily basis. And, yeah, absolutely, that's something that we can do for home automation or our private stuff.

David Beamonte: So, recently, as you have said, as a result of the cooperation between Arduino and Edge Impulse - Edge Impulse is a company that has a platform for AML - we have a way to deploy models on our Arduino boards.

David Beamonte: So, for instance, one of the first tutorials - there are several tutorials that you can use - is to deploy a voice recognition model in an Arduino Nano 33 BLE Sense board, I think it was. So, you have the mic open in the board and it is just listening to the sound around and it can identify certain keywords or certain words according to a specific model that you have trained and you can apply it. So, it follows the traditional way of deploying AML training and deploying.

Peter Dalmaris: So, it could learn my voice, for example?

David Beamonte: Sorry?

Peter Dalmaris: It could learn to recognize my voice.

David Beamonte: Yeah. That's it.

Peter Dalmaris: And I could give a command to turn a switch on, for example. That would be simple.

David Beamonte: Yeah. That's it. That's the point. So, it's kind of you are creating your own kind of Alexa or something similar. So, it depends on the use case that you want. I mean, this is a very simple use case, but you can think about video analytics or other kind of analytics based on inputs. Or you are just reading analog signals, temperature, or humidity, or whatever, and create complex tasks based on that data.

David Beamonte: I think that the limits are very open. So, it's a very, very, very, very vast field of investigation. And I think that, of course, we need to be there. And we are very, very happy to have this integration because this is going to allow our users in a very simple way, as I am always saying, to get started with this.

David Beamonte: I mean, Arduino, apart from education and maker business unit, we also have a Pro view which is more based on industry and so on. And, of course, for that model of business, this is very, very important for factory production and so on.

Peter Dalmaris: It's not just for hobbyists. You're saying that you'll find Arduinos in factories and, probably, supply chains, and supermarkets, perhaps chicken stock. Well, that's very interesting.

Peter Dalmaris: Just going back to machine learning - this is something I'm going to play with - I have tried machine learning platforms elsewhere, but I always find them too complex for me, like in terms of how much time actually I have to put in to learn something new. But with Arduino's reputation of being easy on the beginner, I think there's a good chance that a lot of people would actually finally have a place to try out machine learning and create, I guess, clever, not self-aware, but self-learning applications.

David Beamonte: For me, machine learning has always been things for crazy people just doing lots of algorithms and so on. So, it is true that we need to simplify the message for the majority of the people to let them understand what they can do in their daily basis. And they don't have to learn tense or flow or these kind of things, and learn Python, to learn Python libraries to do that stuff. I mean, we need to simplify things for people who just want to do simple things.

Peter Dalmaris: So, you don't need a PhD to get going with our voice recognition application. Perfect. This is something for 2023 for me that I'm going to play with. And the integration with the ML add-on looks pretty good, which means it's mature for someone new to the space to come in and explore.

Peter Dalmaris: All right. We only have a few minutes left, David, and I still have a lot of questions to ask, but not as important as the things that we have already discussed. So, I'm pretty happy with this. I just wanted maybe to look at the future. I'm not sure how far in the future you can see both yourself as an engineer and as a project manager at Arduino working on Arduino Cloud.

Peter Dalmaris: Is there anything you could tell us that's kind of exciting, both for makers, for teachers, for students, say, students at university now training to be engineers, something that they could be looking forward to from Arduino in the next, say, one or two years? It could be hardware, it could be software, anything you think might be curious and at least very exciting for you.

David Beamonte: Yeah. I mean, when Arduino started, they started with just one board. And if you see now, the ecosystem of hardware and products that we currently have - I don't even think I know all the products that we currently have - we are continuously growing in hardware.

David Beamonte: So, one of the most important things that we do at the Cloud is to try to support everything, support the new hardware that we are producing, new sheets, and so on. But, also, we don't want to force people to use Arduino hardware. I mean, the Arduino Cloud is a platform open for everybody and we are always trying to integrate other platforms. That's why we have ESP boards supported. So, that's something that we will continue doing.

David Beamonte: There are many improvements in terms of widgets, in terms of different stuff, like for instance, creating rules instead of having to create your actions directly from your sketches, that you can program several actions from the Cloud directly. You can just say, "According to these events, I want to do these specific actions and send notifications."

Peter Dalmaris: Like, If, This, and That maybe, so that kind of plug and play. Like, ifthisandthat.com kind of functionality part of the Arduino Cloud. Is that right?

David Beamonte: That's it.

Peter Dalmaris: So, that's coming up?

David Beamonte: Yeah. And, of course, there are more things like plug and play integration with Project Hub. So, you go to Project Hub - for people who don't know what Project Hub is, it's a repository of hundreds, if not thousands, of projects using Arduino boards - you can just with a click integrate the project that you have selected with your board directly. We are trying to improve that journey for beginners mostly, but also for the rest of the people that don't want to spend a lot of time just go [inaudible].

Peter Dalmaris: [Inaudible].

David Beamonte: So, it's very, very convenient. And more integration with the local IDE. Local IDE is the traditional development environment.

Peter Dalmaris: For the new IDE 2.0.

David Beamonte: Yeah, the IDE 2.0. So, for the advanced users, we don't want to force them to use the online IDE, if they feel more comfortable using the local IDE. So, more integration with that. And improvements of the mobile app is also very important.

David Beamonte: There is one very interesting feature that we currently have, which is the mobile app, the mobile phone as a device. So, just with a touch, just with a single step, you can create dashboards to monitor your device, your mobile phone. And then, we are developing another mode, which is a background mode, so that you can just use your mobile phone as another device that is injecting information to the Cloud so that you can use the acceleration or the sound of the mic, or this kind of stuff. So, you can use it for new applications.

Peter Dalmaris: So, you can use the Arduino Cloud without any microcontroller hardware. You can just use your phone as the end device. But I think that because it's got so many sensors in it, you can probably easily combine it with the machine learning add-on and build an application, for example, that would tell you if your walking pattern, for example, is problematic. I guess, you can teach it when you're walking straight and they will tell you like that you need to check your shoes. I don't know, I'm just making things up as I go. Just a rich, rich system of sensors, that's what a mobile phone is.

David Beamonte: Yeah. Imagination is just the limit. So, that's precisely what we want to achieve with this, to invite people and inspire people to create new applications. So, using the phone that can be for many applications. I mean, you can detect accidents, for instance, when you have a very sudden stop or an acceleration or something similar. So, there are many use cases that we can imagine using the phone as an additional device. And interacting with other things at home or in an industry, that there can be many use cases and applications there.

Peter Dalmaris: Perfect. It's another thing to play with. Just now, I'm going to be thinking of applications of the phone on the Arduino Cloud and it could be a good way to explore the AML capabilities now on the Arduino Cloud. So, I'll be thinking about it in 2023 as well. I've got a big list of things that I want to explore in the coming year.

Peter Dalmaris: Okay. David, I guess to to wrap it up, do you have any advice for our listeners and the people watching this? For those people that want to explore IoT, again, not necessarily as a plug to Arduino products, but in general, what would you advise them to do?

Peter Dalmaris: For example, I hear from my students that a big struggle that students have when they start with IoT projects or with programable electronics in general is projects. And that a lot of my students say, "I don't feel that I'm creative enough to come up with an interesting project that I can use to learn from." But there could be other issues that you see that people need help with. So, from your experience, what advice would you give to people listening to this that they want to get into this world?

David Beamonte: I think that people have lots of ideas. So, when they stop thinking about their daily lives, they very easily find things to do. But I think that they struggle to find the way to develop them. So, they don't know how to get started. So, "Okay. I have this idea. I would like to monitor the presence in my corridor and then do something." I don't know, this kind of stuff. But they don't know how to get started.

David Beamonte: So, I think that the first thing that most of the people need to do is to start learning very basic things. Just switch on and off light. Start with very basic projects. Follow some tutorials. And once they become confident with the hardware and how to use it and so on, they will just lose the fear to using that. And then, they will definitely feel very confident to start developing their own stuff.

David Beamonte: So, my suggestion, of course, I'm going to suggest to start using the Arduino Cloud, but there are many other platforms out there. So, it's like following very basic use cases. In our case, in Arduino, we have tutorials from the basic to the complex stuff. So, we are trying to provide users with a journey along their whole learning experience. So, that's my suggestion to start from basic things.

David Beamonte: If they try to start from the complicated stuff at the beginning, they will probably get frustrated. I mean, this is very basic advice. This can apply to everything in life, but also also for this.

Peter Dalmaris: Learn how to walk first - actually, crawl first. Then, you start walking a little, and stable, and perfect walking. Then, start running, and then everything else will follow.

David Beamonte: Crawl, roll, run is coming. Basic life. So, that's my suggestion. And another suggestion would be try to choose just one or two platforms. Because if you want to learn many things, then you are going to get lost. I mean, if that's something that you love, you will definitely be learning many things. But just got to start with a couple of platform. If you love ESP32 board, buy five of them or ten of them. Use it extensively using a specific development framework always, because repetition brings skill and confidence.

Peter Dalmaris: Yeah. Yeah. So, it's important, one thing at a time. You know, when you go on YouTube, you search for one thing. YouTube gives you, like, 20 related things. And then, your problem is, "Which one should I go for?" I find our times are very noisy and that reflects in the difficulty the students have to get started with something.

Peter Dalmaris: Back in my day, I went to school, I had one teacher, one textbook, one path to learning. So, there was not much choice to choose. Things were pretty simple. I look at my kids now, it's just too many roads that they can take to achieve a result, that even that result is kind of fuzzy and not well-defined. So, yeah, you're totally correct and right. Just choose one path and learn one thing at a time. Get it right before you move on to the next thing.

David Beamonte: Yeah.

Peter Dalmaris: All right. Well, thank you, David. I really appreciate your time this morning, for you. I really enjoyed the conversation. I'm going to provide some documentation for our listeners and our viewers to follow through some of the things that we spoke, and hope that they will provide a path for them going forward to discover these amazing technologies that you are also contributing and bringing forth to us. So, thank you.

David Beamonte: Well, thank you for inviting me. And this has been a very cool time. Thank you very much. I enjoyed it very much.

Peter Dalmaris: Awesome.

This is Tech Explorations Podcast episode 16.

Tech Explorations Logo

The Tech Explorations Podcast is a podcast produced by Tech Explorations, a leading provider of educational resources for Makers, STEM students, and teachers. Go to techexplorations.com to see a complete list of our books and courses covering the Arduino, Raspberry Pi, and electronics.


Tags

Arduino, Cloud, IoT


You may also like

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}

For innovators and hobbyists looking to unleash the full potential of their Arduino projects, delving into the realm of wireless technology is essential. Arduino, the compact microcontroller that has revolutionized DIY inventions, can truly flourish

Read More
Exploring Wireless Options for Arduino Projects

I’m excited to introduce you to the latest and greatest version of KiCad – version 8 (stable release). This article will overview the new features and capabilities added to this popular open-source electronic design automation

Read More
KiCad 8: The new and updated features, a full review