1 00:00:00,240 --> 00:00:02,360 Hi, and welcome to the Neil Ashton Podcast. 2 00:00:03,000 --> 00:00:05,880 In each episode, we explain some of the fascinating ways 3 00:00:05,880 --> 00:00:09,080 that science and engineering are changing the world around us. 4 00:00:09,760 --> 00:00:12,720 We talk to leading engineers from elite level sports like 5 00:00:12,800 --> 00:00:16,840 cycling and Formula One to some of the world's top academics to 6 00:00:16,840 --> 00:00:20,445 understand how fluid dynamics, machine learning and 7 00:00:20,445 --> 00:00:22,960 supercomputing are bringing in a new era of discovery. 8 00:00:23,920 --> 00:00:27,040 We also hear some of their life stories, their career advice, 9 00:00:27,640 --> 00:00:30,040 the lessons they've learned on the way that I hope will be 10 00:00:30,040 --> 00:00:33,800 helpful to you too. So sit back and enjoy this 11 00:00:33,800 --> 00:00:40,720 episode. Hi, and welcome back to the Neil 12 00:00:40,720 --> 00:00:43,840 Ashton podcast. Apologies, this episode's a 13 00:00:43,840 --> 00:00:50,400 couple of days late, but I was at the NVIDIA GTC conference in 14 00:00:50,400 --> 00:00:54,560 Washington, DC, and it got me thinking to myself that maybe, 15 00:00:55,320 --> 00:00:59,258 maybe it'd be good to do a quick episode just on the events, 16 00:00:59,258 --> 00:01:03,415 the conferences, the places that I, I would recommend going to 17 00:01:03,415 --> 00:01:07,240 if you wanted, you know, the continuing education to broaden 18 00:01:07,240 --> 00:01:10,160 your horizons to learn more about what's, what's coming 19 00:01:10,160 --> 00:01:13,800 next. Hopefully look at it from the 20 00:01:13,800 --> 00:01:18,200 pure engineering, CAE/CFD sort of angle and maybe also more 21 00:01:18,200 --> 00:01:22,280 on the machine learning angle. And hopefully between the two, 22 00:01:22,280 --> 00:01:26,360 you'll get some good events, maybe some of you listening to 23 00:01:26,360 --> 00:01:28,320 this and go like, I know all this stuff. 24 00:01:28,560 --> 00:01:32,080 I know about where to go to conferences, but I suspect there 25 00:01:32,080 --> 00:01:33,920 are some. And I do get this asked a 26 00:01:33,920 --> 00:01:37,860 reasonable amount of time who yeah, maybe don't quite so 27 00:01:37,860 --> 00:01:40,240 often go to these events. They don't come from that 28 00:01:40,240 --> 00:01:43,880 background and want to know where should they go to get the 29 00:01:43,880 --> 00:01:46,640 latest information. So I'll try and do that. 30 00:01:46,640 --> 00:01:48,190 It'll be a short episode today. 31 00:01:48,198 --> 00:01:51,280 Maybe that's good for some people who find some of the 32 00:01:51,288 --> 00:01:55,030 episodes are two hours long, a bit difficult to get through. 33 00:01:55,038 --> 00:01:59,146 I do put chapters in now, which I'm hoping helps a little bit, 34 00:01:59,154 --> 00:02:01,840 but I know that they can be quite hard to get through in 35 00:02:01,848 --> 00:02:05,219 more in one go. I am trying to work on doing 36 00:02:05,227 --> 00:02:07,932 some short versions of some. So over hopefully the next 37 00:02:07,940 --> 00:02:11,262 couple of months, I'm going to try and put some of the previous 38 00:02:11,262 --> 00:02:14,768 episodes I've got, but put them out as like 10–20-minute 39 00:02:14,776 --> 00:02:16,560 versions. If you don't, you don't have a 40 00:02:16,568 --> 00:02:20,179 chance to watch the whole thing. Listen to the whole thing. 41 00:02:20,187 --> 00:02:24,418 OK, so where would where would I go? 42 00:02:24,426 --> 00:02:29,320 And maybe let's start from the more from a industry point of 43 00:02:29,328 --> 00:02:32,876 view, you know, if you're wanting to really learn about 44 00:02:32,884 --> 00:02:35,652 what's happening in the aerospace industry, what's 45 00:02:35,660 --> 00:02:39,300 happening in space systems, what's happening in aeronautics, 46 00:02:39,300 --> 00:02:44,572 you can't get away from the AIAA events, the American 47 00:02:44,572 --> 00:02:47,776 Institute of Aeronautics and Aeronautics. 48 00:02:47,784 --> 00:02:49,675 Oh, it's aeronautics and astronautics. 49 00:02:49,683 --> 00:02:52,494 I can never remember which way around that is. 50 00:02:52,502 --> 00:02:56,696 These are held every January and they're called, that's called 51 00:02:56,704 --> 00:02:59,340 SciTech and July, which is called aviation. 52 00:02:59,348 --> 00:03:04,761 And these are an interesting event because they're sort of a 53 00:03:04,761 --> 00:03:10,427 bridge between two. You will see C-Suite of major 54 00:03:10,435 --> 00:03:15,045 engineering defense contractors, governments there, you know, 55 00:03:15,053 --> 00:03:20,640 Lockheeds and Raytheons and NASA and others at at those events 56 00:03:20,648 --> 00:03:23,784 doing like very high level. Even some like government 57 00:03:23,792 --> 00:03:26,000 officials will come and do panels, etcetera. 58 00:03:26,008 --> 00:03:31,630 There's an exhibition floor. But how I was always used to the 59 00:03:31,630 --> 00:03:34,103 AIAA is through quite technical talks. 60 00:03:34,111 --> 00:03:40,678 And if you are somebody who works in let's, yeah, let's say 61 00:03:40,678 --> 00:03:44,780 like aircraft design, you know, that's the place to go. 62 00:03:44,788 --> 00:03:47,864 And some of the big workshops there are incredibly useful. 63 00:03:47,872 --> 00:03:50,846 But I'll get to that in a minute. 64 00:03:50,854 --> 00:03:55,940 So you can submit papers, They get published as a conference 65 00:03:55,948 --> 00:03:59,272 paper, you know, very highly regarded in the community. 66 00:03:59,280 --> 00:04:03,280 There is also a journal version of it, Journal of aircraft, 67 00:04:03,288 --> 00:04:07,546 journal of, of other things through the AIAA, you know, not 68 00:04:07,546 --> 00:04:10,892 that cheap to go to. And being completely honest, you 69 00:04:10,892 --> 00:04:13,460 know, sometimes if you're not a student, you're talking 70 00:04:13,460 --> 00:04:16,897 upwards of maybe $700.00 for, for the fee. 71 00:04:16,904 --> 00:04:22,442 I think they've settled now on Florida and San Diego as the 72 00:04:22,450 --> 00:04:25,320 places, but I would still highly recommend. 73 00:04:25,328 --> 00:04:29,608 Fantastic for networking. You will see all of the key 74 00:04:29,616 --> 00:04:33,800 people in the sort of aerospace domain from a CFD and 75 00:04:33,808 --> 00:04:37,852 experimental point of view and some structural work as well. 76 00:04:37,860 --> 00:04:42,002 Not as much on machine learning, but that is starting to change. 77 00:04:42,002 --> 00:04:43,640 There's only one or two sessions 78 00:04:44,680 --> 00:04:48,631 out of you know 10s or hundreds but slowly they are adding 79 00:04:48,631 --> 00:04:50,580 more. There is an exhibition area. 80 00:04:50,588 --> 00:04:54,660 So if your company listening to this, people do sponsor, it's 81 00:04:54,668 --> 00:04:59,320 not that big compared to let's say a supercomputing or other 82 00:04:59,328 --> 00:05:02,674 type of events, but there definitely is sort of commercial 83 00:05:02,674 --> 00:05:05,026 opportunities. I'd say if you're a student, you 84 00:05:05,026 --> 00:05:06,520 should definitely go to one of these. 85 00:05:06,528 --> 00:05:09,900 You know, the ability to bump into people, ask questions. 86 00:05:09,908 --> 00:05:15,769 Technical talks are 30 minutes long, I think, and or 20 to 30 87 00:05:15,777 --> 00:05:18,499 minutes. And the thing that I like the 88 00:05:18,507 --> 00:05:21,770 most about the AIAA are the workshops, things like the 89 00:05:21,778 --> 00:05:25,820 High Lift Prediction Workshop, drag prediction workshop, AIAA 90 00:05:25,820 --> 00:05:30,332 Surrogate Modeling Workshop, which I haven't actually been to 91 00:05:30,332 --> 00:05:34,660 yet, and whatever workshop there was like an acoustics 92 00:05:34,660 --> 00:05:36,680 workshop. I'm forgetting others. 93 00:05:36,720 --> 00:05:39,480 I'm being a bit biased to my area, but lots of workshops that 94 00:05:39,480 --> 00:05:45,080 tend to happen either before or afterwards or sometimes in the 95 00:05:45,080 --> 00:05:49,520 middle depending on the event. And those workshops are, I have 96 00:05:49,520 --> 00:05:52,840 found the best because you basically got all of the things 97 00:05:52,880 --> 00:05:55,320 in one place. They usually book out a room, 98 00:05:55,320 --> 00:05:57,060 you're there for the whole day. 99 00:05:57,068 --> 00:06:01,516 When I think back, you know, 10-15 years ago for me, I met 100 00:06:01,524 --> 00:06:05,708 and learned the most from those, which I really felt what was the 101 00:06:05,708 --> 00:06:08,434 community doing? I just went to some random 102 00:06:08,442 --> 00:06:12,442 talks, not as much, but when it was in the workshop because they 103 00:06:12,442 --> 00:06:14,435 often have like a, a workshop summary. 104 00:06:14,443 --> 00:06:17,700 And that's part of the reason that I interviewed people like 105 00:06:17,708 --> 00:06:20,620 Chris Rumsey in the past, because what they've done for 106 00:06:20,628 --> 00:06:23,680 those workshops, you know, has really been instrumental in 107 00:06:23,688 --> 00:06:27,300 helping people like me and others to understand the area. 108 00:06:27,308 --> 00:06:30,652 And, and that's ultimately why I go to a conference. 109 00:06:30,660 --> 00:06:33,135 I want to make sense of what's the community like? 110 00:06:33,143 --> 00:06:34,980 What's the, what are the challenges? 111 00:06:34,988 --> 00:06:37,990 What's coming next? What are The Who are the main 112 00:06:37,998 --> 00:06:39,716 people? What's the main technologies? 113 00:06:39,724 --> 00:06:43,334 Just get to know people. I mean, now for me, it's friends 114 00:06:43,334 --> 00:06:46,220 as well. It's not just a a work thing, 115 00:06:46,228 --> 00:06:50,040 but I would say you can't beat the AIAA conference if you're 116 00:06:50,048 --> 00:06:52,008 from a government background or like defence. 117 00:06:52,016 --> 00:06:55,400 I mean, you always have to go to those events because the, all 118 00:06:55,408 --> 00:07:00,276 the program managers from the DoD, from the like Air Force go 119 00:07:00,276 --> 00:07:02,650 to those. Again, Russ Cummings, that's why 120 00:07:02,650 --> 00:07:05,147 I interviewed him. You know, he is key part of that 121 00:07:05,147 --> 00:07:06,400 area. You'll probably see him at 122 00:07:06,400 --> 00:07:11,481 almost all the AIAA conferences. So yeah, if you want to know 123 00:07:11,489 --> 00:07:15,904 what's going on in aerospace, whether you're technical or your 124 00:07:15,904 --> 00:07:20,320 sort of C-Suite, I would definitely go to the to the AIAA 125 00:07:20,320 --> 00:07:22,560 conferences. And by the way, this is clearly 126 00:07:22,560 --> 00:07:25,280 not endorsed in any way by any of these conferences. 127 00:07:25,280 --> 00:07:28,000 It's completely just my personal opinion, nothing to do with 128 00:07:28,000 --> 00:07:35,520 NVIDIA or anybody else. I'd say then there are more sort 129 00:07:35,520 --> 00:07:39,884 of niche workshops or events in the aerospace which have to 130 00:07:39,884 --> 00:07:42,878 date normally been done by, by NASA. 131 00:07:42,886 --> 00:07:47,020 NASA has various workshops that they'll tend to do with 132 00:07:47,028 --> 00:07:50,504 aerospace companies from a CFD side, there were things like the 133 00:07:50,504 --> 00:07:53,256 High Fidelity workshop, there was one in May, which was 134 00:07:53,264 --> 00:07:56,298 amazing, really, really good event, had the opportunity to 135 00:07:56,306 --> 00:07:59,909 speak there. And that was all the best sort 136 00:07:59,917 --> 00:08:03,286 of minds and and it was just a great event. 137 00:08:03,294 --> 00:08:06,054 And sometimes they're not that well advertised, so you have to 138 00:08:06,054 --> 00:08:08,595 look for them. But if you were to Google like 139 00:08:08,603 --> 00:08:11,725 NASA Workshop CFD or NASA Workshop Supersonic, you know, 140 00:08:11,733 --> 00:08:16,120 if you're looking for like supersonic aircraft design or, 141 00:08:16,128 --> 00:08:22,164 or hypersonics or something. So yeah, those are very good. 142 00:08:22,172 --> 00:08:28,160 I'm going to ask you to comment maybe if you go, if you're on 143 00:08:28,168 --> 00:08:31,895 YouTube or you can send me a message on LinkedIn, if I post 144 00:08:31,903 --> 00:08:35,010 this on LinkedIn to look at what conferences to go to, because 145 00:08:35,018 --> 00:08:38,503 I'd like to know from you what you think because I maybe I'm 146 00:08:38,510 --> 00:08:42,739 blindsided. There are some in Europe and I'm 147 00:08:42,739 --> 00:08:46,760 forgetting the names of them now, but I have to say, I find 148 00:08:46,768 --> 00:08:50,226 the AIAA ones pretty big. And you know, European folks 149 00:08:50,234 --> 00:08:52,980 from Asia come as well from Australia. 150 00:08:52,988 --> 00:08:58,980 Yeah, can't beat the AIAA. So how about the automotive 151 00:08:58,988 --> 00:09:01,680 side? Well, I don't think there's 152 00:09:01,688 --> 00:09:05,756 anything like the AIAA for automotive to be fair. 153 00:09:05,764 --> 00:09:10,334 Probably the closest would be the SAE, which is the Society of 154 00:09:10,334 --> 00:09:13,232 Automotive Engineers. It's probably the one that has, 155 00:09:13,232 --> 00:09:18,416 yeah, the closest connection, probably a little bit more US 156 00:09:18,424 --> 00:09:21,252 centric and slightly less participation at global level. 157 00:09:21,260 --> 00:09:25,876 I could be wrong, but at least from my opinion, quite a lot of 158 00:09:25,876 --> 00:09:28,960 European folks go to a different one, which is called 159 00:09:28,960 --> 00:09:30,960 FKFS, which I'll talk about in a second. 160 00:09:31,240 --> 00:09:34,920 But if you look at SAE, so you know they have different tracks 161 00:09:34,920 --> 00:09:38,440 and talks, so they have aero, acoustics, aerodynamics, engine 162 00:09:38,440 --> 00:09:42,160 design, car design, systems crash, you know the whole 163 00:09:42,200 --> 00:09:45,800 automotive from a CFD point of view. 164 00:09:45,800 --> 00:09:49,400 There's often talks from the car companies and software 165 00:09:49,400 --> 00:09:53,960 suppliers who who do it. I'd say the technical rigor is 166 00:09:53,960 --> 00:09:57,960 maybe not as deep as at the AIAA, but still still OK. 167 00:09:58,760 --> 00:10:01,480 There is a journal attached to it, although I don't think it's 168 00:10:01,480 --> 00:10:05,280 held in the same regard as the AIAA Journal and personal 169 00:10:05,280 --> 00:10:08,200 opinions here. But still, if you're trying to 170 00:10:08,200 --> 00:10:10,960 understand the automotive community to understand what's 171 00:10:10,960 --> 00:10:14,524 going on, you'll see, you know, some top designers and 172 00:10:14,524 --> 00:10:16,440 engineers from these car companies there. 173 00:10:16,448 --> 00:10:19,656 There was an exhibition area. So for you know, for commercial 174 00:10:19,656 --> 00:10:23,175 companies or start-ups wanting to to go, that's also a good 175 00:10:23,183 --> 00:10:25,320 one. And it's in April in Detroit 176 00:10:25,328 --> 00:10:29,654 every year. So that's one, the European 177 00:10:29,662 --> 00:10:31,600 version. I would say it's probably FKFS, 178 00:10:31,600 --> 00:10:33,545 which is through the Stuttgart wind tunnel. 179 00:10:33,553 --> 00:10:37,076 FKFS, or at least I associate it with the wind tunnel. 180 00:10:37,084 --> 00:10:41,138 Maybe it's a different entity, which is I think every two years 181 00:10:41,138 --> 00:10:45,659 in Germany. And that has, I'd say, you know, 182 00:10:45,659 --> 00:10:50,296 by definition more of a European flavour to it, But 183 00:10:50,296 --> 00:10:54,372 influential people go to that event and has more experimental, 184 00:10:54,372 --> 00:10:56,234 or a mixture of experimental and CFD. 185 00:10:56,234 --> 00:10:59,340 It's got quite a high bar to get accepted there. 186 00:10:59,348 --> 00:11:03,623 So they don't just accept everything and and it's a good 187 00:11:03,631 --> 00:11:05,870 community. So that's probably on the 188 00:11:05,878 --> 00:11:09,617 European angle. There is another one which is a 189 00:11:09,617 --> 00:11:13,380 little bit more now and again, yeah, FKFS is still more 190 00:11:13,388 --> 00:11:17,125 experimental and CFD, whereas CAE is broader. 191 00:11:17,133 --> 00:11:22,075 You can go to AeroVEHICLES. This is a CFD 1. 192 00:11:22,083 --> 00:11:26,127 I think that's also every two years that is experimental in 193 00:11:26,135 --> 00:11:28,136 CFD. It swaps around locations. 194 00:11:28,144 --> 00:11:34,022 That's not a bad one to go to. Now I'm going to plug my own 195 00:11:34,030 --> 00:11:38,230 workshop that I set up with Gary Page and now has got a great 196 00:11:38,238 --> 00:11:40,387 list of people and that's auto CFD. 197 00:11:40,395 --> 00:11:44,508 So the Automotive CFD Prediction workshop, which was basically 198 00:11:44,516 --> 00:11:49,460 inspired by the AIAA workshops that I spoke about, it's 199 00:11:49,468 --> 00:11:53,380 something I created in 2018 off the back of seeing the AIAA and 200 00:11:53,380 --> 00:11:55,400 saying, hey, why isn't there something like this for 201 00:11:55,408 --> 00:11:57,620 automotive? You know, we're always doing 202 00:11:57,628 --> 00:12:01,698 these benchmarks, doing the CFD stuff, but we don't have a place 203 00:12:01,698 --> 00:12:04,544 to go. And so I pitched it for a little 204 00:12:04,544 --> 00:12:08,072 bit to get it done. I was at Oxford at the time, so 205 00:12:08,072 --> 00:12:10,352 we hosted at Oxford. Gary Page at Loughborough, we 206 00:12:10,352 --> 00:12:13,360 did that. And then in the second one, 207 00:12:13,400 --> 00:12:16,560 Urquhart from Ford joined and he's gone from strength to 208 00:12:16,560 --> 00:12:19,480 strength. And so auto CFD 5, it's going to 209 00:12:19,480 --> 00:12:22,160 be next year. So it's usually every two years. 210 00:12:22,200 --> 00:12:27,080 And that is a very focused discussion of CFD methodology. 211 00:12:27,080 --> 00:12:31,480 I would like to say it does have all the main sort of CFD leads 212 00:12:31,480 --> 00:12:34,644 from all the major car companies and software providers now, 213 00:12:34,644 --> 00:12:37,800 but it is very focused on CFD. So if you're interested in the 214 00:12:37,800 --> 00:12:40,160 FEA or anything like that, that's probably not the right 215 00:12:40,160 --> 00:12:45,360 place, but it's I even if I wasn't organizing it, I think it 216 00:12:45,360 --> 00:12:49,440 would be a great event for for you to go to so you can look 217 00:12:49,440 --> 00:12:57,280 that one up. I would then maybe start to move 218 00:12:57,280 --> 00:13:02,212 towards slightly less industrial ones and let's think more 219 00:13:02,212 --> 00:13:08,280 around like the, the topics of the physics or the sub area. 220 00:13:08,760 --> 00:13:14,336 So what I mean by that is you've got things like ETMM, 221 00:13:14,336 --> 00:13:17,716 Engineering Turbulence Modelling and Measurements is a, oh, I'm 222 00:13:17,716 --> 00:13:19,600 sorry, embarrassing here. I know the acronyms. 223 00:13:19,880 --> 00:13:22,600 That's more like a turbulence modelling one, often in a nice 224 00:13:22,600 --> 00:13:25,600 place by the sea. And there's lots of through the 225 00:13:25,600 --> 00:13:28,160 ERCOFTAC. So ERCOFTAC is a European 226 00:13:28,160 --> 00:13:31,080 organization and they host lots of events. 227 00:13:31,360 --> 00:13:33,480 In fact, there's a really good one coming up in March that I'm 228 00:13:33,480 --> 00:13:37,600 helping out with, which is the ML for fluids. 229 00:13:37,720 --> 00:13:40,280 So it's a really nice initiative to try and bring the machine 230 00:13:40,280 --> 00:13:42,960 learning and the fluids communities together. 231 00:13:44,080 --> 00:13:47,800 And that's in Amsterdam. So Richard Dwight is one of the 232 00:13:47,800 --> 00:13:50,880 leading people there, professor there at Delft. 233 00:13:51,760 --> 00:13:54,280 That's another one of these like workshop type ones, which 234 00:13:54,280 --> 00:13:57,132 increasingly I like because they're typically fewer than two 235 00:13:57,132 --> 00:13:58,824 or three hundred hundred people. 236 00:13:58,832 --> 00:14:03,688 They typically have some good invited speakers, really focused 237 00:14:03,688 --> 00:14:06,346 discussions. And considering the theme of 238 00:14:06,354 --> 00:14:09,788 this podcast is often around ML and fluids or CFD. 239 00:14:09,788 --> 00:14:13,840 I would definitely go because it's all about trying to bring 240 00:14:13,840 --> 00:14:15,120 these two communities together. 241 00:14:15,360 --> 00:14:18,920 It's mainly the fluids community, but more and more the 242 00:14:18,920 --> 00:14:23,840 ML people are coming. And so workshops I think are 243 00:14:23,840 --> 00:14:27,160 really good. Now if you are from, let's say 244 00:14:27,160 --> 00:14:31,120 the CFD domain or aerodynamics and you go to the AIAA and you 245 00:14:31,120 --> 00:14:32,960 go to SAE and you go to some of the others. 246 00:14:33,440 --> 00:14:36,880 What you probably aren't aware of unless you've got into this 247 00:14:37,280 --> 00:14:41,840 is the ML conferences. Now the biggest one is NeurIPS 248 00:14:42,120 --> 00:14:47,640 and NeurIPS, and I found out that's how you pronounce it, not 249 00:14:47,640 --> 00:14:50,960 NeurIPS, NeurIPS. There's a reason for that. 250 00:14:51,480 --> 00:14:54,160 That's probably the biggest, most prestigious one where 251 00:14:54,280 --> 00:14:56,320 actually just having a conference paper or even a 252 00:14:56,320 --> 00:14:59,190 poster, it's seen as a big deal. 253 00:14:59,198 --> 00:15:02,884 I've been rejected from that several Times Now, annoyingly, 254 00:15:02,892 --> 00:15:07,500 but I have been and they have lots of workshops as well, like 255 00:15:07,500 --> 00:15:14,005 AI for science, AI for physics. Really good one to go to and I 256 00:15:14,013 --> 00:15:17,280 mean just thousands of people sponsored by all the tech 257 00:15:17,288 --> 00:15:20,255 companies. If you are into machine learning 258 00:15:20,255 --> 00:15:23,166 go, you'll be blown away by the attendance there. 259 00:15:23,174 --> 00:15:27,547 Another one is ICLR, that's another sort of machine learning 260 00:15:27,547 --> 00:15:30,986 one. And the other one is ICML. 261 00:15:30,994 --> 00:15:34,104 So between NeurIPS ICML and ICLR, those are the top ones. 262 00:15:34,112 --> 00:15:39,065 In fact, if you want to get a job at Meta or one of the sort 263 00:15:39,073 --> 00:15:43,590 of tech companies, often you'll see or, you know, NVIDIA in our 264 00:15:43,590 --> 00:15:47,636 like more research labs, you'll see people say must have ICLR or 265 00:15:47,636 --> 00:15:50,600 NeurIPS papers. You know, it's seen as like the 266 00:15:50,600 --> 00:15:53,217 pinnacle from that field. So if you've never been to one 267 00:15:53,225 --> 00:15:54,120 of them, highly recommend you go. 268 00:15:54,120 --> 00:15:57,120 And especially the workshops, I organized one with a few people 269 00:15:57,120 --> 00:16:01,040 ICLR and it was great. You get all people focused in 270 00:16:01,040 --> 00:16:05,000 this room with posters, lots of young researchers as well. 271 00:16:05,800 --> 00:16:08,120 And it's all around the world. They change location every year. 272 00:16:08,120 --> 00:16:10,040 And it's, yeah, just all around the world. 273 00:16:11,880 --> 00:16:17,760 There's a few then that I would say are now going down the more 274 00:16:17,760 --> 00:16:23,040 like commercial route. And this is where, I don't know, 275 00:16:23,040 --> 00:16:26,360 I sort of changed my mind halfway through when I was more 276 00:16:26,360 --> 00:16:28,720 of a pure academic. I would look down upon these 277 00:16:28,720 --> 00:16:31,640 events. So these are the ones like, you 278 00:16:31,640 --> 00:16:34,280 know, your Siemens conference or your Ansys conference or your 279 00:16:34,280 --> 00:16:37,920 Dassault conference in like a vendor conference. 280 00:16:39,160 --> 00:16:42,827 But they are quite good actually from a networking point of 281 00:16:42,827 --> 00:16:47,240 view and from trying to like understand what's going on. 282 00:16:47,240 --> 00:16:51,920 They increasingly now they do try and get some industry 283 00:16:51,920 --> 00:16:53,520 speakers in from different companies. 284 00:16:53,560 --> 00:16:55,080 I've given a talk a few of these. 285 00:16:55,200 --> 00:16:57,220 They'll bring some start-ups in. 286 00:16:57,228 --> 00:17:02,402 They're good if you're more like dealing from a a contracts point 287 00:17:02,402 --> 00:17:05,230 of view or or management point of view. 288 00:17:05,238 --> 00:17:08,781 They're good from a networking and you get to meet a lot of the 289 00:17:08,781 --> 00:17:11,410 key people and they could be a good exposure for your company. 290 00:17:11,410 --> 00:17:13,160 So probably not so much on the 291 00:17:13,160 --> 00:17:17,040 academic side, but they're not bad if your company wants to get 292 00:17:17,040 --> 00:17:20,358 some more exposure and you're thinking from slightly more like 293 00:17:20,358 --> 00:17:24,960 a technical marketing side, the bigger they often do ones now 294 00:17:24,960 --> 00:17:26,839 all around the world. So they'll do like, you know, a 295 00:17:26,839 --> 00:17:30,600 country Pacific version, but they they can be good. 296 00:17:30,600 --> 00:17:34,120 And they often have like workshops to learn the software. 297 00:17:34,120 --> 00:17:36,520 So if you want to get some of your younger staff to go there 298 00:17:36,520 --> 00:17:39,280 and like learn how to use the software, they do that. 299 00:17:39,280 --> 00:17:43,480 So yeah, what I was initially more skeptical of those, but I 300 00:17:43,480 --> 00:17:47,400 think that if you pick the right one, they actually can be not 301 00:17:47,400 --> 00:17:50,000 too bad from a tech point of view. 302 00:17:50,080 --> 00:17:55,440 Then you have my first exposure to this was, well, actually 303 00:17:55,440 --> 00:17:59,600 before we do that, maybe I'd go something like supercomputing. 304 00:18:00,000 --> 00:18:02,400 I always think supercomputing is a good conference to go to, 305 00:18:02,480 --> 00:18:05,538 especially now with convergence of AI and HPC. 306 00:18:05,538 --> 00:18:07,055 I. Wouldn't if you want to 307 00:18:07,063 --> 00:18:09,520 understand the technology, the hardware and the sort of 308 00:18:09,528 --> 00:18:12,560 software level, but maybe a little bit closer to the 309 00:18:12,568 --> 00:18:15,408 hardware side, they're great to go to, very good from a 310 00:18:15,416 --> 00:18:18,500 networking point of view. You know, all the hardware 311 00:18:18,508 --> 00:18:21,371 companies, all the vendors, all the cloud companies. 312 00:18:21,371 --> 00:18:24,600 Increasingly now a lot of the AI companies go to supercomputing. 313 00:18:24,600 --> 00:18:26,280 It gets bigger and bigger and 314 00:18:26,280 --> 00:18:29,440 bigger every year. That's a good one to go to, 315 00:18:29,800 --> 00:18:32,440 especially if you're a company, you're a sponsor or put a booth 316 00:18:32,440 --> 00:18:35,360 there. They have, of course, the Gordon 317 00:18:35,360 --> 00:18:37,000 Bell Prize gets announced there. 318 00:18:37,008 --> 00:18:40,880 So it is quite prestigious. There's lots of 319 00:18:40,880 --> 00:18:43,440 birds-of-a-feather talks, panels, workshops. 320 00:18:43,800 --> 00:18:46,000 So and it's, it's definitely a community vibe. 321 00:18:46,000 --> 00:18:49,520 So it's a bit of a catch all. You know, it it, it's not a 322 00:18:49,560 --> 00:18:52,720 specific discipline per SE. I guess HPC and supercomputing 323 00:18:52,720 --> 00:18:55,880 covers lots of areas, but if you want to learn more about the 324 00:18:55,880 --> 00:18:58,280 hardware, which I think is important, it's part of the 325 00:18:58,280 --> 00:19:02,600 reason I moved to NVIDIA. I wanted to know more about the 326 00:19:02,600 --> 00:19:05,520 the sort of coming from the bottom up where I guess if you 327 00:19:05,520 --> 00:19:07,240 work for software company, you're sort of coming from the 328 00:19:07,240 --> 00:19:09,920 top down. Good event to go to there is ISC 329 00:19:10,000 --> 00:19:12,911 and Supercomputing Asia that are the European and Asian 330 00:19:12,911 --> 00:19:16,430 equivalents to it. Actually I went to the 331 00:19:16,438 --> 00:19:19,568 supercomputing in Europe. It's a much smaller event, but 332 00:19:19,576 --> 00:19:22,948 again, actually pretty well attended and had all most of the 333 00:19:22,948 --> 00:19:25,074 major people around. And you know, it's good to talk 334 00:19:25,074 --> 00:19:26,698 to people and ask them what's their opinion? 335 00:19:26,706 --> 00:19:29,667 What how are things moving? OK. 336 00:19:29,675 --> 00:19:33,306 And what I would mean to say is the tech companies, the my first 337 00:19:33,306 --> 00:19:37,280 exposure to this was re:Invent when I was at AWS and I was 338 00:19:37,288 --> 00:19:40,104 blown away by the size of this event in Las Vegas. 339 00:19:40,112 --> 00:19:44,122 They book out all these hotels and along the strip and there 340 00:19:44,130 --> 00:19:47,588 must be 10s of thousands of people there, just massive. 341 00:19:47,596 --> 00:19:53,020 And I must admit it was really good because they had so many 342 00:19:53,028 --> 00:19:56,810 workshops and in person events where you could go and learn the 343 00:19:56,810 --> 00:19:59,660 software. They had, you know, amazing 344 00:19:59,668 --> 00:20:03,267 keynotes top, you know, figures from biggest companies in the 345 00:20:03,275 --> 00:20:06,242 world. And and yeah, from a networking 346 00:20:06,242 --> 00:20:09,190 point of view, from a learning point of view, you could do 347 00:20:09,198 --> 00:20:12,472 certifications there. And then I saw, you know, 348 00:20:12,480 --> 00:20:15,020 Microsoft, I've never actually been, but Microsoft do 349 00:20:15,028 --> 00:20:18,409 something, Google do something. And you can't help but come away 350 00:20:18,409 --> 00:20:20,960 from these events feeling like, wow, I have just 351 00:20:20,960 --> 00:20:24,000 understood a different part of the tech world. 352 00:20:24,480 --> 00:20:27,760 And yeah, I guess going full circle, I just came back from 353 00:20:27,760 --> 00:20:32,880 NVIDIA's one of these GTC. And, you know, even if I wasn't 354 00:20:32,920 --> 00:20:36,800 working for NVIDIA and when I asked people who have come to 355 00:20:36,800 --> 00:20:39,360 and asked their opinion of it, you know, they were all blown 356 00:20:39,360 --> 00:20:42,760 away and were like, this has helped me so much to understand 357 00:20:42,760 --> 00:20:45,360 the, the tech landscape. I, I sort of understand what's 358 00:20:45,360 --> 00:20:49,000 going on more now, of course you come in there to learn about 359 00:20:49,000 --> 00:20:53,000 that particular company's technology, but you know, even 360 00:20:53,000 --> 00:20:55,200 if you just watch one of these keynotes like like Jensen's 361 00:20:55,200 --> 00:20:58,680 keynote, it, it does sort of open your eyes. 362 00:20:58,680 --> 00:21:01,360 And, and that's certainly how I felt even before I joined. 363 00:21:01,720 --> 00:21:06,920 So I think these events, these industry company events, whilst 364 00:21:06,920 --> 00:21:09,120 you could be cynical and say, oh, they're just like marketing 365 00:21:09,120 --> 00:21:12,760 events actually, especially for the tech companies that, you 366 00:21:12,760 --> 00:21:15,800 know, got so big and have some influence on the world, They 367 00:21:15,800 --> 00:21:20,000 actually, they do give you a lot of insight and it can inspire 368 00:21:20,000 --> 00:21:22,280 you. And yeah, whether you're a 369 00:21:22,280 --> 00:21:25,520 student trying to get a job, go there, network, you can speak to 370 00:21:25,520 --> 00:21:27,200 people. A lot of people come up to me, 371 00:21:27,640 --> 00:21:31,440 you know, who wanted to get some advice on their career or want 372 00:21:31,440 --> 00:21:32,800 to know how to make the first steps. 373 00:21:33,120 --> 00:21:34,800 You know, you can go and speak to these people, you can get 374 00:21:34,800 --> 00:21:37,440 advice. If you e-mail people, are they 375 00:21:37,440 --> 00:21:39,080 going to reply? If you send them a message on 376 00:21:39,080 --> 00:21:42,480 LinkedIn, you know, it's hard. Whereas if you go to an event, 377 00:21:42,480 --> 00:21:46,720 you can go up to them in person and, and get it so, and often 378 00:21:46,720 --> 00:21:48,520 for students, there's discounts, etcetera. 379 00:21:48,520 --> 00:21:52,880 So I, I would, I think I should have gone to more of those 380 00:21:52,880 --> 00:21:55,000 events when I was younger. Actually, I don't think I did 381 00:21:55,000 --> 00:21:58,160 enough of those. And I would, I would definitely 382 00:21:58,160 --> 00:22:02,160 encourage it. So that was, yeah, I probably 383 00:22:02,160 --> 00:22:07,160 missed out so many events and things that I'm probably going 384 00:22:07,160 --> 00:22:10,231 to regret after I posted this. But at least I've given you a 385 00:22:10,239 --> 00:22:13,457 bit of a flavour. Maybe I'll come back to this and 386 00:22:13,457 --> 00:22:16,712 talk about others when I've had more time to think about it, 387 00:22:16,712 --> 00:22:19,608 but as I said, I'll keep this one sweet and short. 388 00:22:19,616 --> 00:22:22,880 And yeah, hope it was useful for you.