Latest / EDGE AI POD / Career EDGE: Navigating the Job Market in the Age of Edge AI

Transcript

Career EDGE: Navigating the Job Market in the Age of Edge AI · EDGE AI POD

• 0:17
  Speaker 1: Hello everyone and welcome to the first edition of the Career Edge Livestream.
• 0:22
  Speaker 1: In this series, we're going to give you no-nonsense, actionable insights to carve a career in the Edge AI industry.
• 0:30
  Speaker 1: Note, even if you're a graduate, early careers professional or experienced professional, there'll be something for you here.
• 0:36
  Speaker 1: I'm Luke Perrins, a specialist talent partner in the industry, and I'll be moderating the panel today.
• 0:43
  Speaker 1: So let's start with an introduction to the speakers.
• 0:47
  Speaker 1: Gregory, let's start with you.
• 0:49
  Speaker 1: Tell us a bit more about who you are, what you do and where you're calling from.
• 0:54
  Speaker 2: Yeah, thank you, luke.
• 0:55
  Speaker 2: How about we start with you actually, because it's your birthday today.
• 0:59
  Speaker 2: So happy birthday, luke.
• 1:01
  Speaker 1: Thank you, Gregory.
• 1:03
  Speaker 2: And before I get started, I wanted to thank the team for putting this webinar together really great organization.
• 1:09
  Speaker 2: I'm really excited to to talk about this topic that's very, very important and very, very dear to my heart.
• 1:15
  Speaker 2: So, um, as you can see on the background, uh, I work for infinium uh technologies.
• 1:21
  Speaker 2: Um, I'm based in austin, texas I know my accent can I give it away a little bit?
• 1:27
  Speaker 2: And I'm the head of product marketing for microcontroller and my team leads the product PSOC Edge.
• 1:33
  Speaker 2: That comes with a neural processing unit and basically a microcontroller dedicated to edge AI applications.
• 1:43
  Speaker 1: Excellent.
• 1:43
  Speaker 1: Thanks a lot, ayman, please.
• 1:47
  Speaker 3: Good evening from Nottingham.
• 1:49
  Speaker 3: It's morning at your side.
• 1:51
  Speaker 3: Good morning.
• 1:52
  Speaker 3: So I'm a professor in pervasive computing and tiny ML at Nottingham Trent University and also honorary visiting professor at Imperial College London.
• 2:03
  Speaker 3: So I work in the area of Tidy Mill Edge AI.
• 2:07
  Speaker 3: I teach, I do research and I engage with the community.
• 2:11
  Speaker 3: I work with many sectors in healthcare, defence, agriculture, bringing the benefits and the applications of Edge AI for a wide range of people and users and sectors.
• 2:28
  Speaker 3: Nice meeting you all.
• 2:31
  Speaker 1: Thanks, Simon and Stephen, please.
• 2:35
  Speaker 4: Sure.
• 2:35
  Speaker 4: Happy birthday, Lou Thank you, you're welcome.
• 2:39
  Speaker 4: I started my career as a technology recruiter and I was doing that for more than 20 years and evolved into being an executive leadership coach in technology for Fortune 500 companies including JP Morgan, Goldman Sachs, Morgan Stanley, a lot of pharmaceutical companies, life science companies, and my job really is to work with senior level professionals all the way down to entry-level people who want to build a career in technology.
• 3:11
  Speaker 4: So my perspective is really from the hiring side, as opposed to someone who's looking for a job.
• 3:17
  Speaker 4: Their perspective is from the job seeking side thanks and Kari, please go ahead.
• 3:25
  Speaker 5: Thank you.
• 3:26
  Speaker 5: I'm a corporate recruiter, an internal recruiter, and I have a wealth of experience in the semiconductor industry and some other industries as well, and there's been a huge increase in use of AI in the semiconductor industry, been a huge increase in use of AI in the semiconductor industry.
• 3:49
  Speaker 5: So I'm looking at it from a perspective of reviewing resumes and, you know, seeking candidates and helping all of you with, you know, with your CV, resume and interviewing and getting seen by people like me.
• 3:59
  Speaker 5: Thanks a lot, carrie.
• 4:00
  Speaker 5: And finally, martin, and interviewing and getting seen by people like me.
• 4:06
  Speaker 1: Thanks a lot, carrie.
• 4:06
  Speaker 6: And finally Martin, yeah thanks guys, good to meet you all.
• 4:09
  Speaker 6: Chief Operating Officer here at WeTech.
• 4:12
  Speaker 6: So we're a startup.
• 4:14
  Speaker 6: We're based in southern Scotland, in Glasgow.
• 4:17
  Speaker 6: So yeah, so we're a startup.
• 4:21
  Speaker 6: My background was very much corporate in terms of of.
• 4:25
  Speaker 6: I was a dialogue semiconductor for some time, but I spent 12 years at Honeywell, so I'm very familiar with both sides of the kind of corporate coin, if you like.
• 4:34
  Speaker 6: So what do we do?
• 4:36
  Speaker 6: We develop power and control system optimization innovations, and we do that at a microcontroller level so effectively.
• 4:45
  Speaker 6: This is pushing AI to the extreme edge, or as we like to call it, the ultra edge, and be very pleased to tell anyone more about ultra edge as we walk through thanks a lot, and thanks to you all for joining today.
• 4:59
  Speaker 1: I really appreciate it for you taking your time out of your of your days to join us and give back to some, you know, graduates in the space.
• 5:07
  Speaker 1: Um, but yeah, I guess, uh, without further ado, let's sort of move on to um.
• 5:13
  Speaker 1: A bit more details then.
• 5:15
  Speaker 1: So I think, um, one thing that would be important to cover is, you know, generative ai typically grabs the headlines in this space, whereas, you know, edge AI sort of works quietly in the background.
• 5:28
  Speaker 1: So I'd like to start with asking you all what edge AI means to you, and we've actually got a really good video from Infineon to start with here.
• 5:37
  Speaker 1: And then, gregory, if you could start with your insights, that'd be great.
• 5:41
  Speaker 1: Yeah, with your insights, that'd be great.
• 6:10
  Speaker 1: Yep, gregory, take the floor.
• 6:18
  Speaker 2: All right, very cool video, yeah, so when we talk about AGI, I think I can relate to what Martin said earlier.
• 6:25
  Speaker 2: For us, it's really about bringing a brand to the device, not to the end device.
• 6:31
  Speaker 2: So basically, it's going to be giving the capability to the device to make decisions, as opposed, as you mentioned earlier, to calling home, basically to have those kinds of information.
• 6:41
  Speaker 2: I don't know if you look at your Alexa, for instance, connected to a cloud and that's where the decision, the AI, is going to be processed.
• 6:48
  Speaker 2: So this brings a lot of benefit.
• 6:50
  Speaker 2: It's going to improve, really, the latency, giving the capability device to make instant decision instead of the back and forth during the communication.
• 6:59
  Speaker 2: It's also an improvement from a power perspective.
• 7:02
  Speaker 2: You're going to be able to not use the high power for the connectivity, and there is a benefit also is, in case you would lose the connectivity, the device is still capable of making those decisions and become very intelligent.
• 7:18
  Speaker 2: We talk a lot about edge AI.
• 7:19
  Speaker 2: I mean for us, it's more like actually edge machine learning.
• 7:22
  Speaker 2: I don't want to go into the controversy and and too deep into this conversation, but it's really what we are talking about here, uh, machine learning being the capability of the device to make decisions and to learn from events and to adapt.
• 7:35
  Speaker 2: So this is really new.
• 7:36
  Speaker 2: That's coming into the picture and I think for us at amphinion we are all about decarbonization and digitalization that that's gonna help also, in the grand scheme of things, to use less power, because you know the big AI cloud server are using a lot of power.
• 7:51
  Speaker 2: So bringing the edge, the AI, to the edge, is gonna actually help with that as well.
• 7:56
  Speaker 2: Maybe one last comment I'd like to share where we're seeing the biggest application right now is in voice, in vision and some predictive maintenance.
• 8:05
  Speaker 2: But to be clear, this is just the beginning.
• 8:07
  Speaker 2: Edge AI is limitless.
• 8:09
  Speaker 2: So we are very, very excited at Infineon to bring this technology.
• 8:13
  Speaker 2: We are a pioneer in this industry with our flagship PSoC Edge microcontroller, which we just showed the video.
• 8:19
  Speaker 2: So that would be my feedback about Edge AI.
• 8:22
  Speaker 1: Thanks a lot, gregory and Kari.
• 8:25
  Speaker 1: Please share your thoughts.
• 8:28
  Speaker 5: So, first of all, recruiting for positions that people have experience in AI and ML so it's very important and it's still niche skills and then we use it for things like helping craft job descriptions, job postings, interview, you know results, all kinds of things recruiting relevant and also letters to candidates you know to make them sound well and things like that candidates you know, to make them sound well and things like that, Okay, Okay.
• 9:09
  Speaker 1: So I guess working quite a lot in your position with the the sort of generative AI side of things as well, obviously you're in people for edge AI positions.
• 9:15
  Speaker 1: Okay, great Thanks for that, and, steven, it'd be great to get a bit more of your, your thoughts.
• 9:22
  Speaker 4: Sure.
• 9:22
  Speaker 4: Thanks for asking To me.
• 9:24
  Speaker 4: Edge AI really refers to implementing artificial intelligence algorithms directly on local devices like either IoT devices or sensors.
• 9:38
  Speaker 4: Literally, they're at the edge of a network.
• 9:41
  Speaker 4: So to me that was always really really impressive because it really allows for real-time data processing and analysis without really relying so much on the cloud infrastructure.
• 9:53
  Speaker 4: So my understanding is it essentially means that processing data is much closer to where it's generated, which really provides faster responses.
• 10:04
  Speaker 4: Reducing latency A lot of what Martin said.
• 10:07
  Speaker 4: I agree with reducing latency compared to sending data to a remote server.
• 10:11
  Speaker 4: My biggest concern about edge AI is there's to me there's a heavy risk of intrusions and security because it's not in a cloud server where you can work with security.
• 10:27
  Speaker 4: But other than that, it's just phenomenal what the capabilities are.
• 10:35
  Speaker 1: It's funny you mention that, Stephen, because it's actually a conversation me and Gregory have had in the past, but I think it opens quite a big can of worms for today.
• 10:47
  Speaker 4: Sorry about that On that point.
• 10:51
  Speaker 1: But yeah, Ayman, please share your thoughts.
• 10:54
  Speaker 3: Thank you.
• 10:54
  Speaker 3: So when it comes to AGI I mean as a researcher I like to go to the other direction.
• 11:02
  Speaker 3: When all the other developers or searchers go into this other direction like, for example, everybody's building these fancy llm models with a lot of parameters you know they need all these computational resources I find an opportunity in democratizing ai, making computing platforms much, much smaller but still with a lot of capabilities.
• 11:27
  Speaker 3: In particular, edge AI can be very beneficial and can change things in terms of privacy.
• 11:33
  Speaker 3: Many people, for example in healthcare, are reluctant to use wearable because they have to share their data with a server, because these wearable have no, they don't have the ability to process the data directly, so they have to rely on a model that sits somewhere else.
• 11:53
  Speaker 3: So you could imagine in terms of healthcare, where people can do self-management, use these wearable without thinking too much about where will their data go.
• 12:04
  Speaker 3: At the same time, hai can be used in other sectors like, for example, farming agriculture, where farmers have very little resources and they don't have enough budgets to buy all these expensive computing platforms.
• 12:21
  Speaker 3: Similarly, in smart spaces in homes and so on.
• 12:27
  Speaker 3: So AGI can open so many opportunities because of the low cost and low energy, but also because of its privacy protection.
• 12:40
  Speaker 3: That appeals to many people, and there are cases where, like for example, if you have this type of wearable that doesn't have any communication tools, you can keep your algorithms to yourself.
• 12:52
  Speaker 3: You don't need even to generate any data, because you get the data and throw it away as soon as you know what kind of insights you get out of it.
• 13:01
  Speaker 3: So there are so many opportunities.
• 13:04
  Speaker 3: We still don't know how far we can go with Edge AI, and that's the excitement for me.
• 13:08
  Speaker 1: Yeah, yeah, for sure, for sure.
• 13:10
  Speaker 1: And finally, Martin, it'd be great to get your insights.
• 13:14
  Speaker 6: Yeah, so I agree with everything that everyone said.
• 13:16
  Speaker 6: You know it is a fantastic opportunity to push the kind of boundaries of autonomous intelligence, but I think we need to remember that it's also kind of an exercise in compromise in terms of we can't do all the great things that we do in cloud at the edge, let alone where we tech operate at the ultra edge.
• 13:37
  Speaker 6: So a lot of what we do is refining, refining, refining and looking at how we can best streamline AI models so that they can run on things like the PSoC Edge that Gregory discussed earlier.
• 13:53
  Speaker 6: So that compromise is something that we've got to bear in mind.
• 13:58
  Speaker 6: So it's not going to do everything that we want it to do, but it needs to be performant and functional for the device that we're connecting to and really nothing more.
• 14:08
  Speaker 6: So that's kind of what edge AI means to me.
• 14:13
  Speaker 1: Okay, thanks a lot.
• 14:14
  Speaker 1: Well, in that case, let's move on to our first section.
• 14:19
  Speaker 1: I'd like to ask you all, in your experiences, what are the most critical technical skills for professionals looking to thrive in the edge AI space?
• 14:30
  Speaker 1: And I think, Martin, why don't we start with you this time?
• 14:35
  Speaker 6: Yeah.
• 14:36
  Speaker 6: So I was thinking about this because you know, when you think about organizations that say you know they use AI in their products, etc.
• 14:46
  Speaker 6: At the edge or anywhere, really, what you need to remember is that it's not just one model, it's not just one technology.
• 14:53
  Speaker 6: It's a collection of many different models and techniques and tools and libraries etc.
• 15:00
  Speaker 6: So really, park in the AI piece for a second.
• 15:05
  Speaker 6: It's not just that you know where we're when, while we're pushing stuff to the edges, that we're connecting to something usually physical.
• 15:14
  Speaker 6: So actually it's a mix in terms of the technical sort of knowledge, really, that we're looking for people in the industry to have embedded software skills, to have you know programming language skills, but equally to have you know insight into hardware, power, electronics in general really, and what the you know the dynamics and theories of all of these.
• 15:38
  Speaker 6: You know quite traditional, in some cases, technologies.
• 15:42
  Speaker 6: Then you've got the data analysis side of things, which is kind of another part that you need to kind of factor in.
• 15:49
  Speaker 6: So you know, once we've got all this data, how do we treat it?
• 15:53
  Speaker 6: Then, coming back to the models, our team use time series models and we use libraries like TensorFlow, lite and Keras, et cetera, which are built on Python.
• 16:06
  Speaker 6: So, all in all, what I would say is that, if you know we were thinking about the skills that you require to be successful in Edge AI.
• 16:18
  Speaker 6: you know, no one's expecting you to have a profound understanding of all of these disciplines, but know all of these disciplines are important and if there's a you know, I guess an area of technology that you want to focus on, focus on that but be aware of the kind of you know the technologies that you're going to need to understand when you get into the practicalities of deploying these solutions.
• 16:43
  Speaker 1: I think that was a really good answer, Martin.
• 16:45
  Speaker 1: Thank you for that.
• 16:46
  Speaker 1: So I guess it's sort of focus on the foundational side of things, but be prepared to develop your skills elsewhere.
• 16:54
  Speaker 6: Absolutely yeah.
• 16:58
  Speaker 1: Yeah, Thanks for that, Martin and Ayman.
• 17:00
  Speaker 1: It'd be great to hear some of your insights.
• 17:03
  Speaker 3: Yeah, sure, I mean as an educator, you know, we always have to answer this question, like, for example, I get master's students who would like to work in this area during their project and they ask me very often like what kind of skills will I gain, what kind of things that I will learn?
• 17:20
  Speaker 3: You know, when I do this project, so I tend to split this into different categories.
• 17:26
  Speaker 3: So, for example, technical skills wise students can learn a lot about how to improve their programming skills, whether it's Python, c++ or other programming languages.
• 17:39
  Speaker 3: Software related skills, because they need to bring and integrate the whole system together, so they need to use software to manage the various operations machine learning operations within that small system.
• 17:51
  Speaker 3: And also they get to learn a lot about embedded systems.
• 17:56
  Speaker 3: And this is where many computing students start to think okay, I have background in ai, I have background in software development, but how about embedded system?
• 18:03
  Speaker 3: I've never used a microcontroller, so I spend some time explaining to them that this is actually still a computing platform.
• 18:12
  Speaker 3: You plug it and you can program and you can upload your code on it.
• 18:16
  Speaker 3: And then, of course, ai skills, where they have not just learn about AI, but also how to do model reduction, how to optimize the AI algorithms so they can fit on these small devices.
• 18:30
  Speaker 3: Besides the technical skills, they have to learn problem solving skills.
• 18:34
  Speaker 3: So they have to understand that this type of model or these type of data can be controlled on this type of device.
• 18:42
  Speaker 3: So you have to come up with a strategy to minimize your model and to come up with different types of data management so you can deploy that model on this device.
• 18:52
  Speaker 3: That is very much constrained in terms of its resources and also they have to be very good in terms of communicating their ideas, like, for example, when we work with group of students.
• 19:05
  Speaker 3: They have to communicate these skills, but also they have to have some extra skills when it comes to deploying these devices in the real world.
• 19:15
  Speaker 3: They turn this into applications.
• 19:16
  Speaker 3: But very often I tell the students that, um, after they finish um their degree, they should have a range of skills that they can present on their cv and they can.
• 19:29
  Speaker 3: That they can also choose what type of career they want, like, for example, if they want to go to ai related industry, if they want to go to embedded and electronic related industry or other type of programming and so on.
• 19:43
  Speaker 3: So they have these varied options and that's what the students like that they are not restricted to one type of industry, but they have many options.
• 19:52
  Speaker 1: Excellent.
• 19:53
  Speaker 1: Thank you, ayman.
• 19:53
  Speaker 1: That's a great answer and I think it's important to touch on the soft skills as well, and I think it links to martin's point quite well that you know you're not going to know all the skills straight away and you know, having a bit of an open mind and a curiosity and willingness to learn new things is going to help you um exceptionally in in this area.
• 20:12
  Speaker 1: Um, carrie, it would be great to get some of your points um on this as well.
• 20:17
  Speaker 1: What have you sort of seen that you know the companies you've worked with have valued the most?
• 20:22
  Speaker 1: Oh, kari, sorry, you're muted.
• 20:30
  Speaker 5: Yeah, your point about enthusiasm for learning.
• 20:33
  Speaker 5: You know that's a big plus and when recruiters or hiring managers see that right on your resume, they like that.
• 20:44
  Speaker 5: Examples of learning you know coding is a hobby, you know things like that.
• 20:50
  Speaker 5: Getting that point across, your enthusiasm for learning more and developing skills you have and learning new skills within the industry we notice those things and mentioning as many things as you can groups that you're in or your projects, your particular contribution to projects, things like that are noticed.
• 21:19
  Speaker 1: Thanks, va Kari and Stephen.
• 21:21
  Speaker 1: It'd be great to hear what your thoughts are.
• 21:24
  Speaker 4: Thanks, luke.
• 21:24
  Speaker 4: Yeah, I really look at this from the perspective of coaching, the executive coaching that I do.
• 21:32
  Speaker 4: I do a lot of work with recent college graduates and even people who've been in application development for a couple of years, taking boot camp courses and, you know, upskilling their knowledge and their capabilities.
• 21:47
  Speaker 4: It's all about because it's such a new area.
• 21:51
  Speaker 4: I think what's really and I agree with what everyone has said up to this point knowledge is important, skill sets important.
• 21:58
  Speaker 4: I think what's critical is understanding how AI frameworks and hardware integration works.
• 22:20
  Speaker 4: Global company and wants to move up into you know, say, from being an architect on the application side or being a developer, and they want to move more into you know AI and, specifically, you know the edge.
• 22:28
  Speaker 4: It's really important to understand how do you select a framework.
• 22:33
  Speaker 4: You know what are the strengths and the weaknesses of popular AI frameworks strengths and the weaknesses of popular AI frameworks.
• 22:41
  Speaker 4: Talking about the users, like, what are users looking for, what kind of documentation is needed and how easy is it to use the selection when you are choosing a framework and then understanding hardware, what are the capabilities of different hardware components that are related to AI, like GPUs and TPUs and FBGAs, which means graphics processing units, tpus or tensor processing units, and FBGAs are field programmable gate arrays, customized hardware that can really be optimized for specific ai algorithms.
• 23:22
  Speaker 4: And then the last part is understanding the integration and the optimization on understanding how to leverage hardware capabilities within the framework that's chosen, using, for example, cuda libraries for NVIDIA GPUs to generate or accelerate computations.
• 23:45
  Speaker 4: So when someone is applying to a position, they don't need to have all the hands-on experience of doing what I just shared.
• 23:54
  Speaker 4: It's understanding the framework, how this is all working together.
• 24:00
  Speaker 4: They all are interoperating together, so that, to me, is extremely important.
• 24:08
  Speaker 1: CHRISTOPHER COUTREAUX-.
• 24:08
  Speaker 1: Thanks a lot, Stephen.
• 24:09
  Speaker 1: And finally, Gregory, great to hear from you as well on this.
• 24:13
  Speaker 2: GREGORY PETERSIER yeah, so I think I will build up on what Martin said.
• 24:17
  Speaker 2: I think we are seeing the same constraints at Infinium.
• 24:19
  Speaker 2: The way I would summarize it is Edge AI is AI in a box, is AI in a very constrained environment.
• 24:28
  Speaker 2: So the challenges for us where we really struggle to find candidates right now is on the hardware side.
• 24:34
  Speaker 2: We need to find people who have experience in NPU and the NPU, the Neural Processing Unit, is new in the embedded world, so our products run with a new 55, down the road U85 from ARM.
• 24:46
  Speaker 2: There's not a lot of experience on that right now.
• 24:48
  Speaker 2: So we have to find people with this kind of experience.
• 24:50
  Speaker 2: On the software side, same thing.
• 24:52
  Speaker 2: You cannot compare a microcontroller with a microprocessor.
• 24:56
  Speaker 2: We have limited resources memory frequency, memory frequency performances.
• 24:59
  Speaker 2: We need to find people who are able to optimize software to make it fit, you know, into the, into our microcontrollers.
• 25:05
  Speaker 2: And also, right now there is a big trend in the micro control industry.
• 25:08
  Speaker 2: People are using a real time os.
• 25:10
  Speaker 2: So expertise in zephyr, expertise in freehouse, are really important for us.
• 25:15
  Speaker 2: We are seeking those kind of skills right now.
• 25:17
  Speaker 2: Then Then on the models we talked about this having experience in programming in Python, for instance, or with TensorFlow, lite, pytorch.
• 25:26
  Speaker 2: All those things are valued on our end and, as we discussed, you know, in the past loop between us, security is important because, in my mind, if you have an expertise in security and how to protect models, the models are becoming IPs for the companies.
• 25:42
  Speaker 2: They're going to spend zillions of money, of dollars, developing those models.
• 25:46
  Speaker 2: They need to be protected because this is IP protection.
• 25:49
  Speaker 2: So, understanding how IPs are going to be protected in the environment and security, those are really the four skills we're really looking for.
• 25:55
  Speaker 2: In addition, if you understand connectivity, that can be a plus Bluetooth, wi-fi, those kind of connectivity that's a plus for us.
• 26:03
  Speaker 2: And, as everybody mentioned, soft skills.
• 26:05
  Speaker 2: Not a single person in the company can know everything.
• 26:09
  Speaker 2: You need to be able to collaborate cross-functionally.
• 26:11
  Speaker 2: So those are the kind of skills that really make a difference and this is really challenging for us.
• 26:16
  Speaker 2: Even if we receive tons of CV constantly, it's really hard for us to find those kind of profiles.
• 26:23
  Speaker 1: Thanks for that, gregory.
• 26:24
  Speaker 1: Yeah, I think everyone sort of agrees on this point, then, that there is a number of higher level skills that are required to have a career in the Edge AI space.
• 26:36
  Speaker 1: But as a graduate, I think you don't need to worry about having all of these skills.
• 26:42
  Speaker 1: As long as you've got a strong foundational knowledge and the willingness and ability to learn, there's going to be opportunities that come your way.
• 26:52
  Speaker 1: So thanks a lot for for all your insights there.
• 26:55
  Speaker 1: I really appreciate that and, um, I guess for our, for our second section then.
• 27:00
  Speaker 1: Um, unless there's anyone that has any other points they wanted to raise on the skills companies value most at all, anything that came to to your heads whilst we were speaking there I have one last comment.
• 27:11
  Speaker 1: Luke if you don't mind real quickly.
• 27:15
  Speaker 4: technology needs to be a solution for business requirements, and everyone's talking about understanding and how important understanding is.
• 27:25
  Speaker 4: However, it's important to look at what the business needs and present AI as the solution to the business, as opposed to promoting AI for the sake of AI.
• 27:38
  Speaker 1: Thanks, thanks for that, stephen.
• 27:39
  Speaker 1: Ok, then, and yeah, let's move on to section two, so how to craft a CV that gets interviews.
• 27:47
  Speaker 1: So I think to give a little bit of insight from me personally on this, you know, first impressions really do mean a lot, and I guess your CV is your opportunity to give that first impression.
• 28:00
  Speaker 1: And I've had it in the past where I've really believed in a candidate and someone that I put forward to a position, but their CV just hasn't really done them justice and I've ended up having to really push for them and they've actually ended up getting the job eventually.
• 28:16
  Speaker 1: So I think it's super, super important to be able to, you know, demonstrate your skills in the best possible way, and it's a super important topic for us to discuss here today.
• 28:27
  Speaker 1: So I guess, stephen, you've got a huge amount of experience with this.
• 28:32
  Speaker 1: What would you sort of say is really crucial when crafting a CV?
• 28:40
  Speaker 4: Well, I'm glad you're asking because you're right, we only have one chance to make the first impression and, because every job description is different, a resume or a CV needs to be edited to match keywords in the job description and, starting as a recruiter and I'm sure you've all had direct or indirect experience interviewing candidates if there is a summary section at the very beginning of a cv or a resume, that's the first chance someone has to really make a strong impression because whoever the reader is whether it's a recruiter or a hiring manager or someone on the technology team if they don't feel that this person is qualified and if they don't feel that the language that's utilized is appropriate, by the time they read 25 to 30% of the page one of the CV, they're going to click off and they're going to look at someone else.
• 29:42
  Speaker 4: So what's really important is to present accomplishments and achievements as they relate to providing solutions for business requirements using artificial intelligence.
• 29:56
  Speaker 4: So talking about specific frameworks like TensorFlow, pytorch, explaining how you understand how they interact with different hardware and providing examples, very specific examples, of how you would optimize models, because it's all about modeling, as everyone keeps saying, it's almost like you know, modeling is proprietary, based on the hardware that's utilized.
• 30:23
  Speaker 4: And, lastly, just highlighting your understanding of factors such as memory management, parallel processing and hardware-specific libraries to achieve efficient AI deployment.
• 30:39
  Speaker 4: That's your shot to make that impression that will motivate the reader to continue looking at the CV.
• 30:47
  Speaker 1: Thanks, Stephen and Ayman, please, it'd be great to get your insights on this.
• 30:52
  Speaker 3: Yeah, as a hiring manager myself what Stephen mentioned I mean sometimes you get so many CVs and many of these CVs have similar kind of skills.
• 31:04
  Speaker 3: So what I look for normally is that the applicant has the required skills but also the ability to improve, to gain further skills, like, for example, working with MPUs, what Gregory mentioned, I mean.
• 31:21
  Speaker 3: These are new types of skills and you will find it very hard to find many graduates with these types of, you know, skills to be able to program and then use MPUs, because these components are coming to the market and they are upgrading every time.
• 31:37
  Speaker 3: So when I I look at CV, I want to have the confidence that this applicant is capable of learning and has the confidence in improving his or her skills, the other things I mean.
• 31:51
  Speaker 3: To get this type of confidence.
• 31:53
  Speaker 3: I would like to see some details, like, for example, adding some metrics on an algorithm that the applicant has developed, like, for example, I improved accuracy by 50% comparing to the traditional algorithms, or I managed to work with this type of hardware and I tweaked it this way and that way.
• 32:14
  Speaker 3: So, given these personalised stories can give the recru recruiter the confidence that this applicant is capable of doing things, but also to improve yeah, yeah, no, I agree with you.
• 32:28
  Speaker 1: I'm in that, and I think one point that that I'd raise, from personal experience as well, is a lot of people, especially when working in a group and I imagine you know at university, when you're working on a group project, you know you do feel like you're working with a group and a lot of people can sometimes put on a resume or CV we, we, we, and I think the most, one of the one key thing as well, is to use the words I, I, I.
• 32:55
  Speaker 1: Otherwise, you know, you can't quite tell how much someone has actually done a certain area of development or whatever it may be.
• 33:04
  Speaker 1: So that's one little point that I'd mention off the back of that, gregory, it'd be great to hear your, your thoughts on this.
• 33:10
  Speaker 1: Gregory, I think you're, you're muted sorry.
• 33:20
  Speaker 2: There's a lot going forward here.
• 33:21
  Speaker 2: I think for us, like it was mentioned earlier, we are really looking at the achievements.
• 33:26
  Speaker 2: We see a lot of candidates that put their position in the company.
• 33:30
  Speaker 2: That doesn't speak a lot to us.
• 33:32
  Speaker 2: To be honest with you, what we are looking for is what have you done that really help us understand?
• 33:37
  Speaker 2: So I think I agree with you about, hey, I've done this and you bring this result.
• 33:43
  Speaker 2: What we've been looking for in resume now more and more is people who are kinda doing some edge AI, being obese with some sort of Arduino or Raspberry Pi, you know what kind of show the passion for the, I would say, this field, those kind of application people who have been creating devices around that, because, like Iman mentioned earlier, it's really hard to find people who have specific end-use skills right now.
• 34:10
  Speaker 2: So those are the kind of that can really make a difference.
• 34:12
  Speaker 2: So if you can highlight in your resume when reaching out for jobs you know we have in Edge AI with what kind of project you've been working that we can relate and can bring value and also really clearly indicate the output and the outcomes on what you're doing in your previous job, that really help us screen faster the candidates In the additional course you could take.
• 34:34
  Speaker 2: You know on AI, you know like the free course you can find on the web from Stanford or even on LinkedIn sometime.
• 34:40
  Speaker 2: Those can bring additional value as well.
• 34:42
  Speaker 2: And maybe one last comment, luke again, if you can highlight cross-functional collaboration work also, that yielded some good results, that makes a difference.
• 34:54
  Speaker 2: Because in big corporations like Infineon we have 60,000 employees.
• 34:57
  Speaker 2: Being able to work cross-functional is really, really critical.
• 35:01
  Speaker 2: So those are things that we value a lot and that can make a difference.
• 35:04
  Speaker 1: Yeah, yeah, yeah.
• 35:06
  Speaker 1: Thanks, gregory and Martin, it'd be great to hear your points on this.
• 35:10
  Speaker 6: Yeah, I mean, it's a different prospect reviewing CVs in a corporate versus a startup.
• 35:16
  Speaker 6: So you know I'm going to ignore the kind of corporate filters and that you shouldn't have a CV that's too visual because it might not make it through the filter system, et cetera.
• 35:27
  Speaker 6: So you know, from a startup perspective, I would certainly review every CV that comes through.
• 35:33
  Speaker 6: So I want the CV to be a reflection of that individual.
• 35:37
  Speaker 6: It gives me a real sort of insight into how they're going to behave and perform when they're in the business.
• 35:45
  Speaker 6: I think that everybody so far has talked about giving evidence of work that they've done, and really I'd interpret that as give me proof, like show me what you've done if you, if you've done it, show me Keeping in your CV things like links to your GitHub repository or you know your blog and your website.
• 36:10
  Speaker 6: We've certainly recruited a couple of engineers who've got, you know, fully-fledged websites and it's just you can see the level of care that they put into these pet projects.
• 36:20
  Speaker 6: You know these personal projects.
• 36:22
  Speaker 6: So you know I've had every faith that when they came into the business they would continue that level of care and they did so.
• 36:31
  Speaker 6: Make your CV stand out visually as well.
• 36:33
  Speaker 6: I'm a visual person so you know if you can create AI, models, code etc.
• 36:39
  Speaker 6: I think you could use something like Canva or Adobe Express to make it look a little bit more engaging than just a plain Word document, and so CV is massively important.
• 36:50
  Speaker 6: But the other thing that's hugely important to me is the cover letter and really emphasizing in that your motivation for wanting to come into the industry and working for know, working for the business, and don't use chat GPT to write it?
• 37:06
  Speaker 1: Yeah, that is actually a question I was going to ask at the end.
• 37:09
  Speaker 1: That is a good point, but just before we get on to that, kari, it'd be great to get your insights as well.
• 37:16
  Speaker 5: Yeah, I mean I agree with everything everyone has said.
• 37:19
  Speaker 5: You know, even if you don't have actual work experience, talking about your personal contribution to projects and showing teamwork with other members working on the project, that is very helpful and, instead of just talking about what the project was, Mentioning things that you've done on the side, like as a hobby or just out of interest, or learning about a particular aspect and other groups that you're in, it shows enthusiasm for learning and it's a chance to also highlight additional things that could be relevant.
• 38:05
  Speaker 5: So and actually we hired someone because they were so impressed with the work that he had done on his own separate so that it can count for a lot.
• 38:19
  Speaker 5: From a more technical point of view, you know, making sure your resume is font friendly for applicant tracking systems is important.
• 38:32
  Speaker 5: Action words and keywords the more keywords that match, the higher up you'll show in searches.
• 38:40
  Speaker 5: Things like that are actually really important when you're looking at hundreds of people applying.
• 38:48
  Speaker 5: So you know things like that as well are important.
• 38:55
  Speaker 1: Yeah, I think that's a really good point you've raised there, kari, about the keywords especially.
• 39:00
  Speaker 1: You know, if you're someone that is looking to apply to the larger multinational businesses, I guess for Martin maybe not so important, probably smaller amounts of applicants important to you know, help you get through these, uh, these ats systems.
• 39:22
  Speaker 1: Yes, um, and yeah, martin, thanks for bringing up the point about the ai and using chat, gpt, um for a cv or resume.
• 39:32
  Speaker 1: It was actually something I wanted to raise myself as well.
• 39:36
  Speaker 1: So I've actually heard sort of two conflicting opinions on this personally.
• 39:43
  Speaker 1: So one view is you know, if someone's used ChatGPT to craft anything in their resume, you know resume CV is straight to the no pile.
• 39:54
  Speaker 1: They're not going to be sort of reviewed Now.
• 39:57
  Speaker 1: The other opposing argument that I've heard of some other businesses is that you know they're just making use of all the tools they have available to them and there's a couple more questions around whether all the skills they've put in their resume are correct.
• 40:15
  Speaker 1: But all it may take is some extra technical grilling in the interview to really be able to work that out, or perhaps an extra stage in the interview process.
• 40:24
  Speaker 1: So I've heard two opposing views myself, I guess.
• 40:28
  Speaker 1: What are all your opinions on it in the call now, I guess, Martin, let's.
• 40:32
  Speaker 1: Let's start with you.
• 40:33
  Speaker 1: It sounded like you were strictly no fair using any AI in this.
• 40:39
  Speaker 6: You can just tell instantly that a cover letter has been written in that, using ChatGPT.
• 40:43
  Speaker 6: It's quite a familiar kind of structure that you get and it's always a little bit over the top in terms of language.
• 40:52
  Speaker 6: I'm thrilled to apply for this fantastic position and blah blah, blah so you know, I don't.
• 40:57
  Speaker 6: I don't want to hear what you know.
• 40:59
  Speaker 6: Some tech company's algorithm, um, you know that spits out in terms of your motivation to join the business.
• 41:07
  Speaker 6: I want to hear from from you, um, and even if the um, you know, the applicant hasn't got english as a first language, I don't care about that, I want to.
• 41:16
  Speaker 6: I want the um, the attitude and attitude and the focus and the ability of that person to make a difference in my business, to stand out, and you don't get that from automated systems 100%.
• 41:30
  Speaker 1: Gregory.
• 41:30
  Speaker 1: Do you have any views on this at all?
• 41:34
  Speaker 2: Yeah, I kind of agree with Martin.
• 41:35
  Speaker 2: It's kind of preferable if we get like a personal touch, you know, and the feel for the candidate.
• 41:42
  Speaker 2: One piece of feedback I would like to share.
• 41:45
  Speaker 2: I see value in chat GPT when it's regarding seeking information about the company.
• 41:51
  Speaker 2: I think too often we show up in a meeting in an interview and we ask hey, tell me about Infineon.
• 41:56
  Speaker 2: Even we are a large corporation, corporation.
• 41:59
  Speaker 2: People don't have really good answer.
• 42:01
  Speaker 2: I would recommend candidate to go and chat gpt and say who is infinium, what are the latest views from infinite.
• 42:06
  Speaker 2: It's good, it's fair game.
• 42:07
  Speaker 2: It's fair game.
• 42:08
  Speaker 2: You show up and you can even say it hey, uh, I've been looking on chat gpt and it indicated you're the worldwide leader in the automotive semiconductor, for instance.
• 42:17
  Speaker 2: It's fair game, you know to say those kind of, those kind of things.
• 42:20
  Speaker 2: So I would say to answer your question in terms of cover letter and resume, we prefer probably the, the personal touch, and not using all those tools.
• 42:29
  Speaker 2: But to prepare for an interview, uh, and and seeking information about the role of the company and being able to talk intelligently about it.
• 42:37
  Speaker 2: I think there is value.
• 42:40
  Speaker 1: For sure, Kari.
• 42:41
  Speaker 1: Do you share the views of Martin and Gregory?
• 42:44
  Speaker 5: I do.
• 42:45
  Speaker 5: We want to see your personality shine through.
• 42:48
  Speaker 5: I think that it's okay to use it to say, put in a paragraph and see what it comes up with, but really it should be your own.
• 42:58
  Speaker 5: It might help add a little finesse to it, but it really should be you.
• 43:03
  Speaker 5: We want to see you and, as someone else mentioned, you know it's quite clear that ChatGPT wrote it and that's an immediate turn off.
• 43:13
  Speaker 5: We want to see your personality shine through.
• 43:17
  Speaker 1: Yeah, yeah, and Ayman, have you got any opinions on this?
• 43:22
  Speaker 3: Yeah, obviously we want to hire people who are smart and resourceful.
• 43:26
  Speaker 3: You know who can use tools.
• 43:27
  Speaker 3: So CharGPT can be used for other things other than you know writing the CV.
• 43:33
  Speaker 3: So, for example, if someone has written their CV and then put it in CharGPbt to ask is there anything missing?
• 43:40
  Speaker 3: Can I add to this?
• 43:42
  Speaker 3: How can I add a wow statement, for example, are there any skills or, like you know, any addition that can improve the CV?
• 43:52
  Speaker 3: So it doesn't need to be written by Chargbt, but it could be, like you know, getting some information from Chargbt to improve the CV and add any missing information.
• 44:03
  Speaker 1: Yeah, yeah.
• 44:04
  Speaker 1: And Stephen, what advice would you give on this?
• 44:07
  Speaker 4: Well, I agree with everything that everyone has said to this point.
• 44:10
  Speaker 4: I'm very pragmatic about cover letters and about resumes and in my opinion, the resume only has one purpose.
• 44:21
  Speaker 4: One purpose and that's to secure an interview.
• 44:25
  Speaker 4: And I always encourage people to write cover letters.
• 44:30
  Speaker 4: And I think the cover letter has two purposes Number one, to demonstrate the person's personality, as Kari has said that's very, very important.
• 44:40
  Speaker 4: And secondly, to encourage the reader to look at the resume.
• 44:46
  Speaker 4: I can't tell you how many cover letters I have seen five paragraphs, six paragraphs, seven paragraphs, and it's always ending with I look forward to hearing from you, which is a mistake.
• 44:57
  Speaker 4: Now, a human being won't even look at a resume unless it gets a high score when it's scanned in an applicant tracking system.
• 45:06
  Speaker 4: If a resume doesn't have between 70 and 80% match in keywords, a recruiter will never even see it.
• 45:14
  Speaker 4: So a lot of people that I talk to, they tell me they use chat GPT to write their resumes and I ask them well, what's your you know, traction?
• 45:24
  Speaker 4: Are you getting interviews?
• 45:25
  Speaker 4: And so many people say, well, I got a couple, but the odds are against me, and that's the reason, because it's a matter of if someone wants to use chat GPT for a cover letter.
• 45:38
  Speaker 4: I think it's a mistake.
• 45:40
  Speaker 4: I think if they want to use it for a sentence maybe they don't know how to write a specific sentence in a resume that might be helpful.
• 45:48
  Speaker 4: But it's very important to edit a document before you hit apply to make sure that there's going to be a high score.
• 45:57
  Speaker 4: And there are programs out there that will allow you to paste your resume and paste the job description.
• 46:04
  Speaker 4: It will scan your resume just like an ATS and give you a score.
• 46:07
  Speaker 4: One of the oldest applications is something called JobScan, which allows you to paste your resume and the job description.
• 46:16
  Speaker 4: So before you apply, you will know that you're getting at least a 70 to an 80% match.
• 46:22
  Speaker 4: And then you're dealing with timing.
• 46:25
  Speaker 4: Has the recruiter already scheduled enough interviews?
• 46:28
  Speaker 4: Maybe they're not looking for any other resumes.
• 46:30
  Speaker 4: Also, maybe enough interviews are happening where offers are being extended.
• 46:36
  Speaker 4: So it's a numbers game and people get discouraged when they get ghosted or when they get an automated response Sorry, job is filled.
• 46:45
  Speaker 4: It's important, especially in technology, to keep that momentum going and use the technology, use the applicant tracking system, to get those interviews.
• 46:57
  Speaker 1: Thank you, Stephen.
• 46:58
  Speaker 1: I think there's a lot of good points that you've raised there.
• 47:00
  Speaker 1: And yeah, before we move on to the next section then does anyone else have any points that they'd like to bring up before we move on?
• 47:10
  Speaker 1: No, All good.
• 47:14
  Speaker 1: Okay, great.
• 47:15
  Speaker 1: So the next section is strategies to stand out and get noticed.
• 47:21
  Speaker 1: So a couple of you have actually touched on this a little bit in the previous section, but I guess we'll focus on it in a little bit more detail now.
• 47:31
  Speaker 1: Martin, let's start with you.
• 47:34
  Speaker 1: What would you say are good strategies people can use to stand out and hopefully land a job?
• 47:41
  Speaker 6: I mean, once you're speaking to a candidate, whether that's on a telephone interview or in a face-to-face interview, whatever it is, I'm really looking for candidates who really imagine doing the job, imagine themselves in the job working in it, what that entails themselves in the job working in it, what that entails kind of thinking through what you'll bring to the job you know, kind of creating, I guess, in their mind.
• 48:07
  Speaker 6: You know what their day-to-day work is going to be like, what use cases that you're keen to work on, you know when they're in the business, and it shows a connection between the candidate and the job in which they're applying for the job in which they're doing, and it gives me confidence that they can come in and, you know, hit the ground running as a startup.
• 48:30
  Speaker 6: The other thing that I'm keen on is for candidates to wear multiple hats where necessary.
• 48:36
  Speaker 6: You know we don't have divisions that do this, this and this.
• 48:39
  Speaker 6: We have individuals that do this, this and this.
• 48:42
  Speaker 6: So if you're going to come into a startup, you need to really understand that at times it's all hands to the pump and you know we have to wear multiple hats to get the job done.
• 48:53
  Speaker 6: So, yeah, so kind of really putting yourself in the, in the, in the role you know.
• 48:58
  Speaker 6: I really thinking that through, think about which role you know and really thinking that through, think about which industries you know that the business can focus on, potentially new industries, potentially new product direction.
• 49:08
  Speaker 6: So you really start to join the dots really, from the technology to delivering the technology and commercialisation, so yeah, that kind of whole piece.
• 49:18
  Speaker 6: Really imagine you're doing the job and convey that to the people that you're speaking to Thanks Martin and Ayman.
• 49:24
  Speaker 1: Do you have any thoughts on this?
• 49:27
  Speaker 3: Yeah, sure, I mean.
• 49:27
  Speaker 3: There are many things that candidates can do to improve their chances, Like, for example, in their CVs they can include any code on GitHub that they have developed, add the link to it to show their practical coding skills.
• 49:44
  Speaker 3: They can, for example, include any research papers they have involved in.
• 49:50
  Speaker 3: Many developers and researchers in AI tend to have some involvement in writing research papers because it shows the performance of the AI models.
• 50:04
  Speaker 3: And the other thing they could promote themselves on social media, For example.
• 50:16
  Speaker 3: Linkedin is a very effective tool.
• 50:18
  Speaker 3: Social media, for example.
• 50:19
  Speaker 3: Linkedin, is a very effective tool.
• 50:20
  Speaker 3: They can write blogs and tutorials on various platforms and they could include these in their social network presence or on CVs.
• 50:34
  Speaker 3: So there are some tangible things that they can show beyond the usual keywords on the CV.
• 50:37
  Speaker 3: So if they include these links to their previous work, it shows that they are capable of doing many things and they have specific skills that they have evidenced in their CVs.
• 50:50
  Speaker 1: Yeah, 100%.
• 50:51
  Speaker 1: Thanks for that, ayman and Kari.
• 50:52
  Speaker 1: What do you think has made people stand out to you and be noticed in the past?
• 50:57
  Speaker 5: I agree links to work that they've done made people stand out to you and be noticed in the past.
• 51:01
  Speaker 5: I agree, links to work that they've done projects they can add also not just on the resume but also on their LinkedIn page, putting those links right in there easy to click on.
• 51:15
  Speaker 5: You can also upload documents to your LinkedIn profile so that people can click on those easily.
• 51:26
  Speaker 5: Joining groups and being active in LinkedIn groups, joining the right kinds of groups you can join quite a lot of them on LinkedIn and taking advantage of that.
• 51:36
  Speaker 5: Also mentioning who your supervising professor is We've searched for candidates specifically from particular professors because they've shown to yet have a very good knowledge of the field that have studied under the particular professor.
• 51:55
  Speaker 5: So things like that also, you know mentorship that you've received or given as well, or things that can help stand out.
• 52:06
  Speaker 1: I think that's a really interesting point about searching on the professors as well.
• 52:10
  Speaker 1: That's a great bit of insightful knowledge there.
• 52:13
  Speaker 1: And also yeah, I think you touched on this a little bit you know networking, joining LinkedIn groups, is super, super important, and I think you know it can go beyond even LinkedIn as well.
• 52:24
  Speaker 1: You know for example, through the EGI Foundation.
• 52:27
  Speaker 1: You know, there's the meetup in Austin coming up.
• 52:29
  Speaker 1: There was other various meetups that took place in 2024.
• 52:33
  Speaker 1: You know, attending these sorts of events, brushing shoulders with people in the industry, communicating with these people, is always going to help you to stand out in the industry.
• 52:45
  Speaker 1: And I guess, Martin, it'd be great to get your opinions on this as well.
• 52:55
  Speaker 2: Yeah, so I agree with everything that has been said here.
• 52:58
  Speaker 2: Maybe I want to add two additional perspectives on this.
• 53:02
  Speaker 2: One thing that's important for us that we look for as well is your approach to different cultures.
• 53:11
  Speaker 2: You know, especially in product marketing, because when we create product, we want to have the understanding.
• 53:16
  Speaker 2: We like to have people who have background, maybe working in different locations, but also having the understanding of different cultures and different needs.
• 53:24
  Speaker 2: So that's one thing that we value people who have been in different places globally, especially for us in product marketing when we create that.
• 53:34
  Speaker 2: Another thing that's pretty important for us is we look also what you do outside of your education and your experience, your work experience, especially in the Edge AI right now.
• 53:47
  Speaker 2: Everybody who has references and I would say, almost hobbies or activity that are very creative, because it's really a field where we're looking for creativity right now.
• 53:56
  Speaker 2: There is so many application that are gonna be invented and created.
• 54:00
  Speaker 2: There is a lot of value for us.
• 54:05
  Speaker 2: So we don't focus necessarily on the education and the experience.
• 54:09
  Speaker 2: We can also look for what you've done outside.
• 54:11
  Speaker 2: Have you been outside of your country?
• 54:14
  Speaker 2: What are you passionate about?
• 54:17
  Speaker 1: Those can make a difference for us in interviews and in the selection process, great thanks a lot, gregory, and a fly distracted me a second ago and I can't remember who I haven't asked the question to yet.
• 54:31
  Speaker 1: So, stephen, great Thanks, stephen, go ahead.
• 54:36
  Speaker 4: Thank you for calling on me.
• 54:37
  Speaker 4: I want to share one word that I have not heard yet, and one of my global clients prioritizes this word together with moving all their applications to the cloud.
• 54:53
  Speaker 4: This is most important to them, and they have 65,000 technologists working for them globally, and the word is innovation, and I agree a lot with what Gregory just shared as far as product marketing, because to me, it's important when making a decision, when interviewing candidates for these types of positions.
• 55:17
  Speaker 4: Hiring a candidate that has the skills and edge AI opens doors to innovation within a company, and anyone that has these skill sets can help drive the development and the delivery of cutting edge technologies, which will probably lead to creating new products or new services or new processes that can set a company apart from their competitors, processes that can set a company apart from their competitors.
• 55:48
  Speaker 4: Innovation is so important.
• 55:49
  Speaker 4: Also, what's very important and I'm asking my fellow panelists, do you agree Decision-making.
• 56:04
  Speaker 4: You know everyone has a different style of interviewing candidates and many of you have hired candidates and many of you interview candidates and there are questions that can be asked like well, you know, tell me about how you optimize efficiencies using edge AI skills.
• 56:16
  Speaker 4: It's something in leadership coaching.
• 56:20
  Speaker 4: That's called case-based interviewing.
• 56:23
  Speaker 4: It's one thing to say, oh, I've done that in every position, every university course I've taken.
• 56:29
  Speaker 4: It's also focusing on optimizing efficiencies.
• 56:33
  Speaker 4: Have examples of how you've saved time.
• 56:37
  Speaker 4: Have examples of how you improved operating efficiencies, whether it's bringing AI capabilities closer to the data source or learning about how you can minimize latency.
• 56:48
  Speaker 4: And also it's all about cost savings.
• 56:52
  Speaker 4: So someone could have all the great skills in edge AI and all the aspirations of building a career in this technology.
• 57:02
  Speaker 4: If it's not cost effective, it's not going to work.
• 57:05
  Speaker 4: That's my two cents.
• 57:09
  Speaker 1: Thank you, stephen.
• 57:09
  Speaker 1: Thanks for that, and does anyone else have any other points that they've thought of, as others have been speaking, or should we move on to the next section?
• 57:18
  Speaker 6: All good.
• 57:19
  Speaker 1: All good, okay, great.
• 57:20
  Speaker 1: So we move on to our fourth and final section then.
• 57:24
  Speaker 1: So how to prepare for an interview and leave a lasting impression, and I think I'm going to split this into two parts.
• 57:34
  Speaker 1: I think, firstly, we'll start with the preparation side of things and then leaving a lasting impression as a second part to this.
• 57:43
  Speaker 1: So let's focus on how to prepare for an interview first and I guess, corrie, let's hear from you please.
• 57:49
  Speaker 5: Right.
• 57:50
  Speaker 5: So I mean using chat GPT is fine for looking up information and news competitors of the company.
• 57:59
  Speaker 5: Research is key.
• 58:01
  Speaker 5: People have gotten jobs or not gotten jobs because either they've researched or haven't, or like we've been really impressed by a candidate that read all of the available technical data that they could, things like that.
• 58:15
  Speaker 5: Looking at chat GPT for possible questions.
• 58:22
  Speaker 5: However, drafting your own responses to these questions is really important and definitely not using chat GPT during an interview to answer questions.
• 58:34
  Speaker 5: So the other thing is thinking of I mean I like the STAR method.
• 58:40
  Speaker 5: You know to give examples what the situation was, the task that had to be done, action and then the result and what your contribution was.
• 58:52
  Speaker 5: Or solving problems, working out, you know, with other people.
• 59:00
  Speaker 5: You know to put forward your ideas and listen to others and what you've learned if things didn't work.
• 59:08
  Speaker 5: Things like that are really important for preparing your answers and having examples easy to hand so that you can remind yourself.
• 59:18
  Speaker 5: You know what your examples are and doing a lot of pre-prep work before any interview.
• 59:27
  Speaker 1: Yeah, thanks for that, kari, you covered it.
• 59:29
  Speaker 1: Yeah, yeah, great points, great points.
• 59:31
  Speaker 1: And Ayman, what do you think on this?
• 59:35
  Speaker 3: So the beauty of Edge AI is that you can build a small device and you can take that device with you wherever you go.
• 59:44
  Speaker 3: So I always advise my students and graduates, if they can, to build a demonstration of their work or, if they have done work on hardware and AI, to make it more like a presentable demo and take it to the interview, because the interviewers can relate more when they see some examples and they can talk about it.
• 1:00:08
  Speaker 3: The candidates can talk about how they build it and show this example live.
• 1:00:14
  Speaker 3: So that would be a very interesting approach to take your project with you.
• 1:00:20
  Speaker 1: Yeah, I really like that idea.
• 1:00:21
  Speaker 1: That's a great idea and anything else, simon, at all that you sort of advise your students on.
• 1:00:31
  Speaker 3: So, in general, students with coding skills in medicine tend to be shy a little bit.
• 1:00:38
  Speaker 3: So showing some confidence and social skills.
• 1:00:42
  Speaker 3: You know we tend to spend a lot of time behind the screen coding and you know sometimes it could be very hard to go to the real world and interact and be friendly enough.
• 1:00:53
  Speaker 3: You know we are friendly, but it's different from other disciplines.
• 1:00:58
  Speaker 3: So I would like I also advise that students and graduates try to spend more time presenting themselves and being able to hold a conversation in a friendly way and show their personality as well as well as their coding and technical skills, because many graduates tend to talk about the technicality a lot and they go head on towards the technicality, but it's good also to talk about other things.
• 1:01:24
  Speaker 1: Yeah, yeah for sure, I think you know it's natural to feel anxious, especially if it's good also to talk about other things.
• 1:01:27
  Speaker 1: Yeah, yeah, for sure.
• 1:01:27
  Speaker 1: I think you know it's natural to feel anxious, especially if it's, you know, your first big interview after leaving university, and I think you know no interview is ever going to be there to catch someone out.
• 1:01:40
  Speaker 1: So it's just about you know going in there, you know trying not to be so nervous about it, and just you know getting across who you are and what you can do.
• 1:01:49
  Speaker 1: And as long as you manage to do that, you know I'm sure you'll be fine.
• 1:01:53
  Speaker 1: There's definitely no tricks or traps that anyone's trying to set up for anyone at all.
• 1:01:58
  Speaker 3: Thanks for that.
• 1:01:59
  Speaker 3: There are times where there are tricks.
• 1:02:08
  Speaker 6: Oh, have you got any of those tricks up your sleeve.
• 1:02:10
  Speaker 6: I've got loads of tricks Pulling someone's chair out from under their nose.
• 1:02:15
  Speaker 6: Yeah, I think from a preparation perspective, obviously you have to research the customer, know the company inside out, but also research the competition.
• 1:02:28
  Speaker 6: You know we're especially.
• 1:02:29
  Speaker 6: You know if you're a commercial business you're in business not necessarily just to build great tech.
• 1:02:35
  Speaker 6: You know you're in business to sell great tech.
• 1:02:37
  Speaker 6: So you know how do we position ourselves against the competition.
• 1:02:41
  Speaker 6: You know who are we seeing out there that we can kind of compare ourselves with.
• 1:02:45
  Speaker 6: So it's not just researching the company, research the competition, particularly interested when people come up with new potential customers as well.
• 1:02:53
  Speaker 6: You know applications of the technology.
• 1:02:55
  Speaker 6: So even if you're just reading a white paper that waxes lyrical about technology and what the tech does, take that into the real world and say, actually, you know, I think that could really work well in the railways and I've got experience of doing a project and some kind of rail transportation system where this could really apply really well.
• 1:03:16
  Speaker 6: It might not be something that we've thought about and that really would strike me as somebody who's really getting the technology and also really thinking about the real world applications really.
• 1:03:28
  Speaker 6: So that's what I would do to prep.
• 1:03:31
  Speaker 1: Alright, thanks for that, Martin and Stephen.
• 1:03:33
  Speaker 1: Let's hear your thoughts on this.
• 1:03:37
  Speaker 4: Thanks for asking.
• 1:03:38
  Speaker 4: In my opinion, there's only two things that need to happen for someone to move to the next round and to get an offer.
• 1:03:45
  Speaker 4: Number one, communicating that the person has the skills, the knowledge and the experience to do the job as it's written.
• 1:03:55
  Speaker 4: And number two, that the person's communication, personality and behavior will fit in the culture of the company, and candidates applicants won't know the culture of the company and candidates applicants won't know the culture of the company.
• 1:04:09
  Speaker 4: They really won't.
• 1:04:10
  Speaker 4: They can do all the research in the world going on Glassdoor and other websites, but because we really only have one chance to make that first impression, it's important to be concise and direct and answering questions, and usually you know, unfortunately there's no university course on how interviewers conduct interviews, so you know everyone kind of does it with their own style and their own methodology.
• 1:04:37
  Speaker 4: Almost everyone that I know of will end an interview with thanks for answering all my questions.
• 1:04:43
  Speaker 4: Luke, do you have any for me?
• 1:04:45
  Speaker 4: With thanks for answering all my questions, luke, do you have any for me?
• 1:04:50
  Speaker 4: Most people that I work with they ask the wrong questions.
• 1:04:51
  Speaker 4: It's like what keeps you up at night.
• 1:04:53
  Speaker 4: You know it's the last chance someone has to reinforce their candidacy, to get to the next round and to receive the offer.
• 1:05:02
  Speaker 4: It's got to be thought provoking.
• 1:05:04
  Speaker 4: It must really push the buttons of the interviewers by asking questions like the person that's lucky enough to get this great opportunity with your company after they've been working with you for three months.
• 1:05:18
  Speaker 4: What performance criteria, what metrics will you analyze to feel confident that you offered the job to the best candidate?
• 1:05:25
  Speaker 4: To feel confident that you offered the job to the best candidate?
• 1:05:30
  Speaker 4: Don't just ask run-of-the-mill questions.
• 1:05:34
  Speaker 1: That's my two cents.
• 1:05:35
  Speaker 1: No, I think that's brilliant, stephen, thanks for that.
• 1:05:37
  Speaker 1: And finally, gregory, let's hear from you on this one.
• 1:05:41
  Speaker 2: Yeah, I think Stephen is bringing a good point here.
• 1:05:44
  Speaker 2: The last question is really important.
• 1:05:47
  Speaker 2: Don't say I don't have any questions, that's it.
• 1:05:52
  Speaker 2: But I kind of like what Steven said.
• 1:05:54
  Speaker 2: I kind of like to be challenged.
• 1:05:55
  Speaker 2: So it can be like can you tell me what the KPIs are going to be for that role?
• 1:05:58
  Speaker 2: Define success for that role?
• 1:06:10
  Speaker 2: One I like as well is when somebody tells me what is your vision in this market and how do you think infinity superior to competition?
• 1:06:13
  Speaker 2: Asking what's the vision?
• 1:06:14
  Speaker 2: Because if you're going to call me to the company, you want to understand where we're going and if you are really understand the vision, if you agree with this vision.
• 1:06:17
  Speaker 2: So I think that's where the the rules are being reversed.
• 1:06:20
  Speaker 2: Uh, we're challenging you for the job, but now you can challenge.
• 1:06:23
  Speaker 2: Challenge us to see if we are the right fit for you.
• 1:06:27
  Speaker 2: So those are the kind of questions I like.
• 1:06:30
  Speaker 2: I would say know the company, know the role, get prepared on.
• 1:06:35
  Speaker 2: This is really important.
• 1:06:36
  Speaker 2: It's really important for us.
• 1:06:38
  Speaker 2: Just to conclude on that I want people to understand that we are just as anxious as the candidate to find the right candidate.
• 1:06:49
  Speaker 2: We are also stressed when I show up in an interview.
• 1:06:51
  Speaker 2: I mean, this is important.
• 1:06:53
  Speaker 2: We have a very you have, for instance, a very well-oiled machine, you know, as a team that's working.
• 1:06:59
  Speaker 2: You're going to bring one more individual into the team.
• 1:07:01
  Speaker 2: That might be disruption.
• 1:07:02
  Speaker 2: That might be disruptive.
• 1:07:04
  Speaker 2: So, finding the right candidate that's going to perform in this very competitive environment, this is a challenge for us as well.
• 1:07:09
  Speaker 2: We're just as stressed.
• 1:07:10
  Speaker 2: So just be yourself.
• 1:07:12
  Speaker 2: It benefits everybody at the end of the day.
• 1:07:14
  Speaker 2: To be honest and just to be yourself luke.
• 1:07:18
  Speaker 4: Can I just say once real quick, gregory, gregory, what you're saying resonates with me so much because everyone that's looking for a job thinks it's about them.
• 1:07:27
  Speaker 4: I have to plan what I'm going to say.
• 1:07:31
  Speaker 4: I have to plan how I'm going to answer questions.
• 1:07:34
  Speaker 4: What's really helpful is if a candidate looks at it from the hiring side.
• 1:07:39
  Speaker 4: Gregory, when you schedule an interview, you have nerves.
• 1:07:43
  Speaker 4: You have the job description in front of you, you have questions that you're going to be asking.
• 1:07:48
  Speaker 4: You know what answers you want to be hearing from each candidate.
• 1:07:52
  Speaker 4: So I think it's that perspective.
• 1:07:56
  Speaker 4: It's not just about the person looking, it's about what are the expectations from the interviewer?
• 1:08:04
  Speaker 1: Yeah, yeah, and I've actually question um for for all of you guys and girls actually.
• 1:08:11
  Speaker 1: So, um, one question that I've I've previously given advice to candidates um in the past, especially in a final stage interview is you know, clearly, some of the questions that you brought up, gregory, you know around sort of like questions that you brought up, gregory, you know around sort of like progression, company vision, very, very important as well.
• 1:08:31
  Speaker 1: But also there's there's quite an uncomfortable question to ask um and that's, you know, at the very end of all the questioning.
• 1:08:38
  Speaker 1: You know, do you have any um reservations about my ability to to work on this in this position and do this role?
• 1:08:48
  Speaker 1: What do you all feel about asking that sort of question?
• 1:08:54
  Speaker 1: I guess, from my perspective, it gives the ability to be able to react to any reservations within the interview and make sure all questions are answered and there's a full understanding of someone's skill set.
• 1:09:05
  Speaker 1: But I guess it'd be good to understand your opinions on that.
• 1:09:10
  Speaker 4: I think that's negative.
• 1:09:12
  Speaker 1: No, really.
• 1:09:13
  Speaker 4: Yes, really.
• 1:09:15
  Speaker 1: And why do you think that, Stephen?
• 1:09:19
  Speaker 4: I would much.
• 1:09:19
  Speaker 4: Rather well, it's almost like asking why am I not qualified?
• 1:09:24
  Speaker 4: That's why, instead of asking the question the way you framed it, Luke, what if you were to say based on everything that you've heard and everything that you've read on my resume, can I ask you, do you have any further questions?
• 1:09:41
  Speaker 4: The candidate will know by the body language if it's positive or negative.
• 1:09:49
  Speaker 1: Yeah, yeah, no, I would agree with that.
• 1:09:50
  Speaker 1: But yeah, I guess I think you never want to leave the final stage interview with any questions left unanswered.
• 1:09:58
  Speaker 1: So yeah, I think I'd agree with you.
• 1:10:00
  Speaker 1: Stephen, the way you position the question was definitely, you know, a better way to position it compared to how it's not a contest.
• 1:10:08
  Speaker 1: It's not, yeah, no no, no, no, no, no.
• 1:10:10
  Speaker 1: I know that, yeah, no, but yeah, definitely the way you sort of position that question is a better way to answer the ask the question.
• 1:10:17
  Speaker 1: But I think it is important, in whatever way you do it, to make sure that you don't leave the final stage interview with any stones left unturned because ultimately you know they probably aren't going to come.
• 1:10:29
  Speaker 1: They're going to be interviewing a number of other candidates and they're probably not going to come back to you and ask you a question about something they felt you were missing.
• 1:10:38
  Speaker 4: So yeah.
• 1:10:38
  Speaker 1: I think, yeah, I think we're sort of in agreement now.
• 1:10:41
  Speaker 1: I just positioned the question in the wrong way.
• 1:10:44
  Speaker 1: But yeah, no thanks for that, stephen.
• 1:10:46
  Speaker 1: Thank you for that.
• 1:10:47
  Speaker 1: No, my pleasure.
• 1:10:48
  Speaker 4: Luke.
• 1:10:48
  Speaker 4: One last thing.
• 1:10:49
  Speaker 4: Most interviewers will say thanks, luke, someone will get back to you.
• 1:10:54
  Speaker 4: Don't ever leave the interview that way.
• 1:10:57
  Speaker 4: Someone may not get back to you.
• 1:11:01
  Speaker 4: So one of my candidates when I was recruiting said this to me and it opened my eyes and I love everyone's opinion about this.
• 1:11:08
  Speaker 4: Thanks for taking the time.
• 1:11:10
  Speaker 4: It was a pleasure to speak with you.
• 1:11:12
  Speaker 4: In case I'm close to receiving an offer somewhere else, would you like me to notify you?
• 1:11:26
  Speaker 1: It's creating a sense of urgency because you know that additional interviews are being scheduled for this position and you know, if you were saying, oh, we'll get back to you, yeah, yeah, yeah for sure.
• 1:11:32
  Speaker 1: And what do you think about?
• 1:11:33
  Speaker 1: You know like thank you emails after an interview process.
• 1:11:38
  Speaker 1: Do you ever recommend that to people you're working with?
• 1:11:41
  Speaker 4: All the time, but it's much more than thank you.
• 1:11:43
  Speaker 4: It's thank you and, as we discussed, you need A, b, c, d and E, and as we reviewed my background, it sounded like you're on the same page as I am.
• 1:11:55
  Speaker 4: I've done A, b, c, d and E, and then end it with I'm looking at a number of other opportunities.
• 1:12:02
  Speaker 4: However, your company and this position appears to be the best match for my skill sets and my career goals, and don't end it with.
• 1:12:11
  Speaker 4: I look forward to hearing from you yeah, take control of the situation.
• 1:12:16
  Speaker 2: Yes, sir, yeah, yeah if I can build up on that.
• 1:12:19
  Speaker 2: Look for us in marketing, 100 percent, 100 percent.
• 1:12:22
  Speaker 2: We need to follow up email because this is how we behave with customer.
• 1:12:26
  Speaker 2: You go to the customer, you take notes.
• 1:12:28
  Speaker 2: After the meeting, you send notes, say, hey, here is the notes I took during the meeting, here's what matters to you, here's the action items I noticed.
• 1:12:35
  Speaker 2: We will get back to you in a couple of by middle of next week, something like that.
• 1:12:39
  Speaker 2: Perfect.
• 1:12:40
  Speaker 2: So if you don't see this professionalism in the person, especially in marketing, I think for more technical role, that can be okay, but for marketing roles, you know, or customer facing roles, if the candidate doesn't send a follow up email.
• 1:12:56
  Speaker 1: That's a big negative for us.
• 1:12:57
  Speaker 1: Okay well, thank you guys.
• 1:12:57
  Speaker 1: I think we're sort of coming close to the end of time now so we better have a quick look at the audience questions.
• 1:13:05
  Speaker 1: But thanks again for all of your insights there.
• 1:13:09
  Speaker 1: So just going to take a quick look at some of the questions we've got in the comments section, bear with me, I've never been close to the end of time.
• 1:13:19
  Speaker 4: This is really Okay.
• 1:13:26
  Speaker 1: So we've got a question here from Deborah and she asks with AI becoming more and more evolved, do you think it will eventually become less clear that AI has written a personal statement or a cover letter?
• 1:13:43
  Speaker 1: Do you think it will be harder to distinguish between the two?
• 1:13:46
  Speaker 1: And I'll just keep this open to anyone that wants to answer it.
• 1:13:50
  Speaker 1: We'll just have sort of one answer to these questions to save time.
• 1:13:54
  Speaker 3: I would say, yes, we already see this.
• 1:13:57
  Speaker 3: You can't change the style of what you're writing, so it will be harder and harder to distinguish.
• 1:14:03
  Speaker 3: You can't tell language models to change their styles and to be more human-like.
• 1:14:09
  Speaker 3: You can take the text and improve it a little bit, so we're already seeing evidence of that.
• 1:14:15
  Speaker 3: So more advancement in that direction will make it even harder for humans to detect that this is written by a machine.
• 1:14:24
  Speaker 1: Yeah, it's only going to get more harder and harder to distinguish between the two, I'm sure, and we've got a question here from John Banks.
• 1:14:33
  Speaker 1: So he asked what's the current state of the job market when hiring an international applicant as an engineer with two years of experience?
• 1:14:44
  Speaker 1: He's had a bit of tough luck applying to quite a few jobs and not managed to land an interview.
• 1:14:50
  Speaker 1: Um, so I'm not quite sure what the sort of what's the current difference?
• 1:14:56
  Speaker 1: Yeah, so I mean, does anyone have any sort of opinions on on how to answer that question?
• 1:15:03
  Speaker 6: I can only say from my perspective, we do get quite a lot of applicants where we would need to sponsor that applicant and really for us it's a cost that we can't really justify at the moment as a startup.
• 1:15:16
  Speaker 6: We have to kind of look at the whole picture and look at the market locally and see if there's a talent there.
• 1:15:21
  Speaker 6: I think as we grow you that's something that we would do, but you know, potentially it's adding a level of complexity to the research that personally, as a startup, we can't afford from a time perspective or a cost perspective.
• 1:15:39
  Speaker 6: However, when we have graduates working on a graduate visa within the UK in particular, we absolutely want to keep those people within our business and we will go through that sort of sponsorship approach.
• 1:15:55
  Speaker 6: But yeah, I think it will differ to which companies you apply for.
• 1:16:00
  Speaker 6: It's just for us that would be a no at the moment, just because there is talent here and it's an extra time, extra cost and time for it and any sort of points from those of you that have worked in larger multinational environments at the moment.
• 1:16:19
  Speaker 2: So maybe I can comment on that.
• 1:16:20
  Speaker 2: I agree with Martin it's getting harder and harder to have those visa sponsorship.
• 1:16:23
  Speaker 2: The different approach at Infineon is we're getting harder and harder to have those visa sponsorship.
• 1:16:26
  Speaker 2: The different approach at Infineon is we're global, so we are more open about the location of the candidates.
• 1:16:33
  Speaker 2: If we have the right candidates we can be more flexible because we have offices everywhere.
• 1:16:37
  Speaker 2: But I agree it's generally very, very hard to get sponsorship right now, generally very very hard to get sponsorship right now.
• 1:16:47
  Speaker 3: I could add from university perspectives.
• 1:16:52
  Speaker 3: Sometimes we have this opportunity but it depends on the qualifications of the applicants.
• 1:16:58
  Speaker 3: So, for example, if they have a master's degree or PhD, then they can apply for research assistant or research fellow and these can be open for international applicants.
• 1:17:13
  Speaker 1: But it depends on the qualifications and the skills.
• 1:17:15
  Speaker 1: Okay, thank you for that and, stephen, do you have any opinions on this one at all?
• 1:17:24
  Speaker 4: I think the only feedback I have really Luke is applicant tracking systems.
• 1:17:27
  Speaker 4: If he's not getting traction, maybe that's the reason.
• 1:17:30
  Speaker 1: Yeah, good point.
• 1:17:31
  Speaker 1: Yeah, john, I mean, if you are watching now or if you see this back, feel free to share your resume with myself or Stephen, and we can take a look at that to see if there's any advice we can give you.
• 1:17:45
  Speaker 1: Okay, then I'll just get a couple more questions before we finish.
• 1:17:50
  Speaker 1: So please give some examples on what kind of new positions will emerge in the edge AI sector.
• 1:17:59
  Speaker 1: Quite an interesting question, so I'll leave it to the floor.
• 1:18:06
  Speaker 6: I would say don't think of it as just engineering roles.
• 1:18:08
  Speaker 6: These are complex, potentially complex projects which need to be rolled out.
• 1:18:13
  Speaker 6: Therefore, having a good program and project managers with experience in image AI maybe transitioning from things like IoT-type projects into these kind of spaces is going to be something good.
• 1:18:30
  Speaker 6: Also, I mean I'm more on the commercial side and we're looking for really good salespeople to take this message to the world as well.
• 1:18:40
  Speaker 6: So the whole kind of gambit really around what you need to build a business is going to be required for a GI.
• 1:18:48
  Speaker 6: So don't just think it as engineering roles.
• 1:18:54
  Speaker 1: Does anyone?
• 1:18:54
  Speaker 1: Else have anything to add?
• 1:18:55
  Speaker 2: Yeah, I would like to share some feedback that I had from my colleague at Imagimob.
• 1:19:00
  Speaker 2: Imagimob is a company that was acquired by Infineon to create models.
• 1:19:04
  Speaker 2: There is new opportunities there they talk about when I asked them.
• 1:19:10
  Speaker 2: They talk about finding people who are able to really get the data to create the models.
• 1:19:15
  Speaker 2: This is really challenging.
• 1:19:17
  Speaker 2: Those are new job actually to identify people who will be able to collect those data.
• 1:19:20
  Speaker 2: So, because most of the device naturally don't necessarily transmit those data, but if you don't have those data you cannot create those models and the value of the model highly depend on the data you can correct.
• 1:19:34
  Speaker 2: So those are kind of the new jobs we're finding and also, on the model side as well, being able to optimize those models, because we are talking about edge AI here and, like I said earlier, this is AI in a box, so you cannot say I'm just going to run those PyTorch and TensorFlow models.
• 1:19:53
  Speaker 2: They have to be optimized.
• 1:19:54
  Speaker 2: We don't have those gigabytes of memory available.
• 1:19:57
  Speaker 2: Everything is constrained.
• 1:19:58
  Speaker 2: So people who are able to optimize those models, there is a lot of value in the industry right now to optimize the cost.
• 1:20:04
  Speaker 2: So I think that's where I see a lot of opportunity On industry right now to optimize the cost.
• 1:20:07
  Speaker 2: So I think that's where I see a lot of opportunity.
• 1:20:09
  Speaker 2: On the sales and marketing front, I would say being able to explain the benefits and bring really the piece of the of the npu, the, the ai, explain to the customer how they can differentiate in their industry by leveraging edge AI and be able to promote a complete solution to the customer.
• 1:20:28
  Speaker 2: Because a lot of customers are kind of scared of it.
• 1:20:30
  Speaker 2: I thought they don't really need to use it.
• 1:20:32
  Speaker 2: Those bring new opportunities and new skills that we'll be on the lookout for.
• 1:20:38
  Speaker 1: Thanks, gregorio, and does anyone else have any thoughts on this?
• 1:20:42
  Speaker 4: I noticed a few jobs Principal AI engineer doing edge AI, face recognition.
• 1:20:48
  Speaker 4: There are jobs in research, generative AI and edge AI research and development, senior edge AI system engineers.
• 1:20:58
  Speaker 4: There's a lot of different roles Data scientists, artificial intelligence engineers, deep learning engineers, machine learning engineers Because this technology is continuing to evolve and based on where a company is, where their products are and where their future lies.
• 1:21:20
  Speaker 4: It's someone's ability to fill a job that requires AI as a solution to their business requirements.
• 1:21:27
  Speaker 4: There's a ton of opportunities in research, which really isn't technology directly, but also in the engineering side, because there are so many engineers already, I just think there are going to be more and more job opportunities in engineering.
• 1:21:45
  Speaker 1: Thanks for that, stephen, and does anyone else have any final points on that question?
• 1:21:49
  Speaker 1: No, Okay.
• 1:21:53
  Speaker 1: So Andreas has asked what do you think would be the outstanding curriculum to start studying now to be a successful applicant in five years' time?
• 1:22:04
  Speaker 1: It has been IT 10 years years ago and it has been machine learning engineer two or three years ago.
• 1:22:10
  Speaker 1: If your kid was to start studying now, what would you recommend?
• 1:22:17
  Speaker 3: great question yeah, very good question obviously I mean you could start with the solid math skills.
• 1:22:26
  Speaker 3: Hai in optimizing AI algorithm require a lot of skills in math and being able to understand the depth of AI model, how to deal with it, how to quantize it, how to reduce the size of that model and so on.
• 1:22:48
  Speaker 3: So this requires really good understanding and good foundations in math, and then the other technical skills like programming as well, the ability to deal with electronics without the feeling that it's a difficult subject and I can't work with solder electronics on a circuit and so on.
• 1:23:13
  Speaker 3: So I think being comfortable with embedded systems, with math, with AI, at early stage could help in terms of preparing someone to create and invent in the edge AI world.
• 1:23:27
  Speaker 3: But I mean in five years' time.
• 1:23:29
  Speaker 3: We don't know what will happen right In five years' time.
• 1:23:32
  Speaker 3: It might not be even edge or anything.
• 1:23:33
  Speaker 3: It could be a smart dust or so everything will be embedded in the real world.
• 1:23:41
  Speaker 3: Contextual computing will take over as well, and where everything can detect everything and then respond to it.
• 1:23:48
  Speaker 3: So we will not see even HAI in sight.
• 1:23:53
  Speaker 4: However, you did say math, which is so important.
• 1:23:56
  Speaker 4: Sorry to interrupt Math.
• 1:23:58
  Speaker 4: I think you're right.
• 1:23:59
  Speaker 4: It's such an important skill set early on, and also data management, I think.
• 1:24:06
  Speaker 3: Yeah.
• 1:24:06
  Speaker 1: I think that's a super hard prediction to make on that question for sure.
• 1:24:10
  Speaker 1: But yeah, interesting Even physics.
• 1:24:12
  Speaker 3: Physics can help as well.
• 1:24:15
  Speaker 3: There are a lot of techniques that are physics-inspired, even brain-inspired techniques, if you understand how brain works.
• 1:24:22
  Speaker 3: There's a lot of work in neuromorphic computing at the moment, and a lot of people try to understand how the brain works.
• 1:24:29
  Speaker 3: How can we mimic the human brain and and, um you know, use it in the design of hardware, for for ai or ai for hardware?
• 1:24:37
  Speaker 3: So, um, these skills can be very, very helpful.
• 1:24:41
  Speaker 3: Like I find myself reading about brain all the time, so I wish if I have these skills when I was little me too okay, and then I think we'll do this.
• 1:24:53
  Speaker 1: Uh, one final question and then wrap things up.
• 1:24:55
  Speaker 1: So if, if I am an embedded engineer and have never worked in the ai space before, but want to break into ai, what are the top two skills, um, that you think someone would need to learn, uh, to actually, you know, transition across to a business working in the edge ai space I would say sorry.
• 1:25:18
  Speaker 6: I would say it's less about uh, skills, it's kind of what are the experiences?
• 1:25:24
  Speaker 6: So, if you want to break into aji, break into aji, you know, build some projects.
• 1:25:30
  Speaker 6: Um, you know, in your, in your personal time, um, really kind of show that you've got that understanding.
• 1:25:36
  Speaker 6: The thing is, um, you know the, the models and the tools, etc.
• 1:25:41
  Speaker 6: These will all change.
• 1:25:43
  Speaker 6: You know they'll change over time.
• 1:25:44
  Speaker 6: So don't get too invested in any one technology.
• 1:25:47
  Speaker 6: Learn a you know a breadth of the technology and then really focus on what you want, which industry you want to go into, which, what really excites you and, um, you know, from our perspective, um, that's looking at motor control and power inverters.
• 1:26:02
  Speaker 6: So, industrial automation, automotive, you know, really kind of get a real good understanding of the industries that you want to go to and understand what's important to those industries and really understand the company that you want to apply for.
• 1:26:17
  Speaker 6: You know what's their focus.
• 1:26:20
  Speaker 6: Somebody like Infineon has lots of different focuses, you might find that there's, you know, another semiconductor company out there that's really really focused on automotive or domestic appliances, et cetera.
• 1:26:33
  Speaker 6: So focus your learning, focus your knowledge development on where you want to be.
• 1:26:40
  Speaker 4: That's a great answer.
• 1:26:41
  Speaker 4: I was just going to say get a certification as well.
• 1:26:44
  Speaker 4: Intel has an Edge AI certification.
• 1:26:47
  Speaker 4: I think that would help.
• 1:26:51
  Speaker 1: Great, thank you and anyone else with any other points or two.
• 1:26:55
  Speaker 2: Well, the two that comes to mind for me, if I was to answer that question right now, would be developing the capability to integrate, to optimize models and integrate them in your software project, in your IDE.
• 1:27:09
  Speaker 2: How to cause that's a skill we're looking for, how to really integrate those models when you create a project and, of course, we're creating the tools to make it easier.
• 1:27:17
  Speaker 2: So that's one that I like to see.
• 1:27:20
  Speaker 2: I mean, just like Martin was saying, if you've done some side project related to that and you can show you know how to do that, that's a great skill.
• 1:27:26
  Speaker 2: But also the embedded engineering in terms of software right now is really need towards real time operating system.
• 1:27:35
  Speaker 2: So we tend to value people who have knowledge of FreeRTOS or Zephyr.
• 1:27:39
  Speaker 2: So those would be the recommendation I would have.
• 1:27:43
  Speaker 2: Those are the kind of skills we are looking for right now.
• 1:27:46
  Speaker 2: I'm sure there is plenty more, but those are the two that came to mind when I saw the question.
• 1:28:01
  Speaker 1: Great.
• 1:28:01
  Speaker 1: Thanks a lot, gregory.
• 1:28:01
  Speaker 1: Well, look, I think that's all the topics for today.
• 1:28:02
  Speaker 1: I'd like to thank everyone for the questions and to all the panelists, thank you for taking the time out to join us.
• 1:28:07
  Speaker 1: We'll be releasing everything covered today in an Edge AI career guide, the first volume of the career guide.
• 1:28:16
  Speaker 1: After this live stream, it'll be available on a Discord channel.
• 1:28:22
  Speaker 1: But, yeah, thank you for watching and hope to see you all again soon.
• 1:28:26
  Speaker 1: Thank you, thank you, channel um.
• 1:28:27
  Speaker 1: But, yeah, thank you for watching and, uh, hope to see you all again soon.
• 1:28:28
  Speaker 1: Thank you, thank you.
• 1:28:28
  Speaker 1: Oh, we're still alive, yeah, the pub isn't open yet.