1 00:00:00,160 --> 00:00:03,600 Today I'm joined by Perus Pulatrac, Software Engineer over 2 00:00:03,600 --> 00:00:06,200 at booking.com and apparently he's a living legend. 3 00:00:06,280 --> 00:00:09,120 If you ask anyone about hackathons within Booking, 4 00:00:09,400 --> 00:00:11,880 they'll point you to him. He's actually won a couple or 5 00:00:11,880 --> 00:00:15,440 so, and with close to 1000 engineers participating, that's 6 00:00:15,440 --> 00:00:18,080 quite the accomplishment. It's one of his main passions 7 00:00:18,080 --> 00:00:21,480 and he's going to explain why hackathons are valuable for you 8 00:00:21,480 --> 00:00:24,720 as software engineers and for a company to organize. 9 00:00:25,080 --> 00:00:28,680 They even organize hackathons in their graduate programs and 10 00:00:28,680 --> 00:00:30,600 there's a lot of skills you get from participating. 11 00:00:30,640 --> 00:00:33,400 After this conversation, I'm sure you'll want to participate 12 00:00:33,680 --> 00:00:35,200 as well. So enjoy. 13 00:00:35,200 --> 00:00:42,080 Do you do that often? Go from an idea and kind of a 14 00:00:42,080 --> 00:00:45,240 need where you see there's no real good solution. 15 00:00:45,240 --> 00:00:47,920 Or even when it's just an experiment, you create a tool 16 00:00:47,920 --> 00:00:51,800 out of that. This is kind of my hobby, So 17 00:00:52,440 --> 00:00:56,320 what I'm doing is like each time I have some sort of idea or 18 00:00:56,320 --> 00:00:59,640 problem in my life, it doesn't matter what it is, I'm always 19 00:00:59,640 --> 00:01:03,360 looking how can I like solve this problem for myself and also 20 00:01:03,400 --> 00:01:07,240 anyone has the similar problem. So I did it for multiple 21 00:01:07,240 --> 00:01:08,720 applications. Some of them are already 22 00:01:08,720 --> 00:01:13,200 published, some of them I, I didn't publish yet and like it. 23 00:01:13,280 --> 00:01:16,000 It could be like, for example, I created something for 24 00:01:16,640 --> 00:01:19,440 appointment Hunter and this is what this is like. 25 00:01:19,640 --> 00:01:22,120 Getting appointment for some government was really 26 00:01:22,120 --> 00:01:25,400 challenging for me and I've seen people struggling and you need 27 00:01:25,400 --> 00:01:28,040 to go to the page, refresh and refresh and so on. 28 00:01:28,040 --> 00:01:31,280 So I said why we spending a lot of time doing that? 29 00:01:31,280 --> 00:01:35,000 We can just automate it. So I did it in like even less 30 00:01:35,000 --> 00:01:38,360 than a less than an hour. I create an application, it goes 31 00:01:38,360 --> 00:01:41,280 and does everything for you, sending notification when it's 32 00:01:41,280 --> 00:01:43,640 done. So you can do the same thing. 33 00:01:43,880 --> 00:01:47,880 You can enable yourself for the job of work that is like you're 34 00:01:47,880 --> 00:01:50,680 struggling now. And I shared this application 35 00:01:50,800 --> 00:01:53,920 and apparently that was super interesting for the HR of the 36 00:01:53,920 --> 00:01:57,040 company I was working and they share it with all people and 37 00:01:57,040 --> 00:01:59,840 they were really happy. I got a lot of feedback, but 38 00:01:59,840 --> 00:02:02,600 that was like hundreds line of JavaScript code. 39 00:02:02,600 --> 00:02:05,720 So there's nothing really challenging, but I could do it 40 00:02:05,720 --> 00:02:09,080 like in an. So this is my hobby to be 41 00:02:09,080 --> 00:02:11,240 honest. I even develop games, I develop 42 00:02:11,240 --> 00:02:14,840 some application for myself, develop even for different 43 00:02:14,840 --> 00:02:18,760 technology, not not for the web itself, like using Electron, I 44 00:02:18,760 --> 00:02:22,760 develop application for Macintosh and like other OS. 45 00:02:23,320 --> 00:02:26,000 And that's really feasible. Like if you know the ground 46 00:02:26,000 --> 00:02:28,640 knowledge, you can easily, there's no need to go to 47 00:02:28,640 --> 00:02:32,560 documentation, right? Like before and go and read a 48 00:02:32,560 --> 00:02:34,760 lot of documentation, understand all aspects. 49 00:02:34,760 --> 00:02:37,880 You need to know the general capabilities of technologies. 50 00:02:38,320 --> 00:02:41,520 Then you can easily puzzle that up and make what you want. 51 00:02:42,200 --> 00:02:45,800 I feel like building and creating applications regardless 52 00:02:45,800 --> 00:02:49,160 of the platform, even seeing problems and then trying to find 53 00:02:49,160 --> 00:02:51,480 a solution in in software, it's a real skill. 54 00:02:51,960 --> 00:02:53,720 And I'm wondering from your perspective, since you say this 55 00:02:53,720 --> 00:02:56,080 is a hobby of mine, how do you become good at that? 56 00:02:57,320 --> 00:03:01,240 I think the best thing to being a good at this skill is doing 57 00:03:01,240 --> 00:03:02,880 it. So it's like you need to try, 58 00:03:02,880 --> 00:03:05,640 you need to fail, you need to like do it more and more. 59 00:03:06,080 --> 00:03:10,000 And I always like, I remember the first, first like 60 00:03:10,000 --> 00:03:12,280 application I was started, I struggled a lot. 61 00:03:12,280 --> 00:03:16,520 It was not really working well. But I even started this hobby 62 00:03:16,520 --> 00:03:20,240 before this AI era. So that was more like 63 00:03:20,240 --> 00:03:24,640 challenging and time consuming for me from the start of this 64 00:03:24,640 --> 00:03:27,640 era. It was like a really happy time 65 00:03:27,640 --> 00:03:31,000 for me to be honest because I jumped into the project in a 66 00:03:31,000 --> 00:03:33,640 day. So I even created one like 67 00:03:33,920 --> 00:03:35,880 optimization application for myself. 68 00:03:35,920 --> 00:03:40,480 I have these printers that you have in the stores, thermal, 69 00:03:40,880 --> 00:03:44,240 thermal printers. And for my daily work, it 70 00:03:44,400 --> 00:03:46,800 automatically prints out my to do tasks. 71 00:03:46,800 --> 00:03:50,640 So I created with JavaScript connecting with some server and 72 00:03:50,640 --> 00:03:54,440 web hook and just that's patched everything like in a couple 73 00:03:54,440 --> 00:03:58,360 hours and I have like a printer everyday prints out my To Do 74 00:03:58,360 --> 00:04:01,040 List. This is like really like like 75 00:04:01,200 --> 00:04:05,680 old fashioned way of doing that. But I want to just experiment 76 00:04:05,680 --> 00:04:10,040 doing different stuff, making some crazy and that like this AI 77 00:04:10,040 --> 00:04:12,640 tools enabled me to do it super fast. 78 00:04:12,640 --> 00:04:17,000 So I did it like so many things like this before, but it took me 79 00:04:17,120 --> 00:04:21,560 a month or two months sometimes to and like your vibe is not 80 00:04:21,560 --> 00:04:23,720 always you cannot do it really fast. 81 00:04:23,720 --> 00:04:27,000 But now when I start, most of the time I finish the same day. 82 00:04:27,480 --> 00:04:32,040 So it, it doesn't like take more than one day for me to getting 83 00:04:32,040 --> 00:04:33,880 something. The production ready stayed. 84 00:04:34,320 --> 00:04:36,720 But the code, most of the code is right only. 85 00:04:36,720 --> 00:04:43,600 So it's not that high quality so but it works with some because 86 00:04:43,600 --> 00:04:45,800 I'm not going to sell those product. 87 00:04:45,800 --> 00:04:49,320 But this is like MVP. But it it enable me and resolve 88 00:04:49,320 --> 00:04:53,680 my problem, I can go ahead. I also tried to make one of the 89 00:04:53,680 --> 00:04:58,720 projects in high quality, so that took more time, but I could 90 00:04:58,720 --> 00:05:02,040 break it down with the help of AI with the standard I know like 91 00:05:02,520 --> 00:05:05,360 each function should be really short, like file name, 92 00:05:05,400 --> 00:05:07,680 everything. I could also do it the same way, 93 00:05:07,680 --> 00:05:12,720 but it takes way more time and you should start it really soon. 94 00:05:13,000 --> 00:05:16,280 You should not like make the product working then switching 95 00:05:16,280 --> 00:05:18,360 to change it. So it's really horrible. 96 00:05:18,840 --> 00:05:22,160 You need to start from the ground, don't allow the AI go 97 00:05:22,160 --> 00:05:25,000 the wrong way. Pluck it and go step by step. 98 00:05:25,000 --> 00:05:27,880 It takes more time but the end result can be more production 99 00:05:27,880 --> 00:05:30,240 ready. If you want to sell it, if you 100 00:05:30,240 --> 00:05:33,520 want to have something that that is maintainable later. 101 00:05:34,400 --> 00:05:37,080 But this product, most of the product I generate is not really 102 00:05:37,080 --> 00:05:41,440 maintainable, but it does the job I want and that's enough for 103 00:05:41,440 --> 00:05:45,440 me. There's no like security really 104 00:05:45,440 --> 00:05:49,680 layer in some of them and some of them which need security, I 105 00:05:49,680 --> 00:05:51,720 have even four or five layer of security. 106 00:05:51,760 --> 00:05:55,480 Because if if you know it and you know, you see this is like a 107 00:05:55,480 --> 00:06:00,040 challenge for the security can already ask AI to do that or you 108 00:06:00,040 --> 00:06:02,920 need to do it manually. Sometimes it's not always good 109 00:06:02,920 --> 00:06:05,000 at doing everything to get the security. 110 00:06:05,240 --> 00:06:08,400 Yeah, that makes sense. I mean, this the skill of 111 00:06:08,400 --> 00:06:11,760 getting something up and running in one day or even a few days, 112 00:06:12,080 --> 00:06:14,800 it really reminds me of hackathons where there's a 113 00:06:14,800 --> 00:06:16,400 competition and this is the goal. 114 00:06:16,400 --> 00:06:19,600 Basically everyone comes together either as individuals 115 00:06:19,600 --> 00:06:23,240 or as groups that just have formed that meet this goal that 116 00:06:23,240 --> 00:06:25,560 they have set, and then they have a few days to execute on 117 00:06:25,560 --> 00:06:27,160 it. What has been your experience 118 00:06:27,160 --> 00:06:29,680 with hackathons? Exactly, I'm passionate about 119 00:06:29,680 --> 00:06:32,880 hackathons and in my company I'm like a symbol of hackathon. 120 00:06:33,520 --> 00:06:37,160 You're the man. And yeah, like every time 121 00:06:37,160 --> 00:06:39,960 speaking with people, they start talking about hackathons. 122 00:06:40,520 --> 00:06:45,320 I even like the one that going out from the like representative 123 00:06:45,320 --> 00:06:47,920 of the company going out and speaking about hackathon 124 00:06:47,960 --> 00:06:51,280 universities as well, because this is like really my really 125 00:06:51,280 --> 00:06:54,360 real passion. So that's the same same reason 126 00:06:54,360 --> 00:06:58,680 because this comes from my hobby that is generating something and 127 00:06:58,680 --> 00:07:02,120 like ideating, solving the problem and that goes to the 128 00:07:02,120 --> 00:07:05,720 hackathon as well. And I found like the opportunity 129 00:07:05,720 --> 00:07:10,360 in my company also to having these availability to having 2 * 130 00:07:10,360 --> 00:07:13,160 a year hackathons. And I've done a lot of crazy 131 00:07:13,160 --> 00:07:17,840 project in hackathons, which I'm really proud of because the end 132 00:07:17,840 --> 00:07:22,240 result was not only in the hackathon, I could like get 133 00:07:22,240 --> 00:07:28,320 something out to the world and like get real result out of 134 00:07:28,320 --> 00:07:30,480 that. And that was really interesting. 135 00:07:30,960 --> 00:07:33,720 I didn't know that it's really even possible in the past that 136 00:07:33,720 --> 00:07:36,840 you can do something hackathon and at the end you get something 137 00:07:37,280 --> 00:07:40,160 real life out there. You have an example of what you 138 00:07:40,160 --> 00:07:42,200 built in the hackathon and what made it in the product. 139 00:07:42,600 --> 00:07:44,720 Exactly, I have quite a few example. 140 00:07:45,120 --> 00:07:47,840 I can start with the one of them, which was the start of the 141 00:07:47,840 --> 00:07:52,840 ChatGPT era. We, we formed A-Team to create 142 00:07:52,840 --> 00:07:55,680 the agent. So our mindset was to having 143 00:07:56,720 --> 00:08:00,640 something like before that we had some people, you go there 144 00:08:00,640 --> 00:08:03,200 and just they manage everything for, for your trip. 145 00:08:03,760 --> 00:08:06,280 I wanted to create the same thing with the AI. 146 00:08:06,920 --> 00:08:11,280 That was really like crazy idea at that time because our company 147 00:08:11,280 --> 00:08:14,880 like booking.com is huge. So it's you cannot easily attach 148 00:08:14,880 --> 00:08:17,360 everything to the like simple agent. 149 00:08:17,920 --> 00:08:20,320 So we started with the attraction only. 150 00:08:20,320 --> 00:08:23,840 So to look into the attraction, we created the hackathon and we 151 00:08:23,840 --> 00:08:27,120 started working on that. And what we did was to create an 152 00:08:27,120 --> 00:08:32,200 agent that you can chat and it owes already encrypted with the 153 00:08:32,200 --> 00:08:35,320 generative UI. And it could generate UI for the 154 00:08:35,320 --> 00:08:37,559 list and show you show you a list of item. 155 00:08:37,640 --> 00:08:41,880 Then you can buy or select, add to your like selections or save 156 00:08:41,880 --> 00:08:46,560 it as like a bookmark and so on. Then you can ask question about 157 00:08:46,560 --> 00:08:49,920 your whole trip. So for example, say you are in 158 00:08:49,920 --> 00:08:53,120 Paris and you want to do all activities, restaurant and some 159 00:08:53,120 --> 00:08:57,600 of them even not available at booking.com but you want to plan 160 00:08:57,600 --> 00:08:59,720 it, you want to have itinerary for yourself. 161 00:09:00,040 --> 00:09:02,640 So the idea was to discuss, ask questions. 162 00:09:02,640 --> 00:09:06,600 So you can say I have like with two children, a dog and like 163 00:09:06,600 --> 00:09:11,240 these circumstances, it can be even complex circumstances and 164 00:09:11,240 --> 00:09:14,520 it's not even easy for human or everyone to manage that. 165 00:09:14,520 --> 00:09:19,080 But AI can easily break down. And also we created something 166 00:09:19,080 --> 00:09:22,920 like vector database to search through all availability that we 167 00:09:22,920 --> 00:09:26,040 have and serve better result for people. 168 00:09:26,520 --> 00:09:30,880 It was just for the attraction. And that was quite nice 169 00:09:30,880 --> 00:09:34,120 experience for me that we developed something crazy. 170 00:09:34,120 --> 00:09:39,200 There was even no, that time I remember Open AI, there was like 171 00:09:39,200 --> 00:09:44,000 API for open AI, but the ChatGPT quality was different between 172 00:09:44,480 --> 00:09:49,280 different with the API that we had available globally. 173 00:09:49,280 --> 00:09:53,880 So what we did was to creating a wrapper over the ChatGPT and 174 00:09:53,880 --> 00:09:57,360 like a hacking, hacking way of getting access to ChatGPT 175 00:09:57,360 --> 00:10:01,480 because the quality was better. We wanted a really high quality, 176 00:10:01,640 --> 00:10:05,040 high quality LLM to analyse everything that we had because 177 00:10:05,040 --> 00:10:08,440 hallucination was really a thing that time and now it's way 178 00:10:08,440 --> 00:10:12,160 better than before. Then we even learn prompt 179 00:10:12,160 --> 00:10:13,760 engineering throughout the process. 180 00:10:13,760 --> 00:10:19,360 So the, the, IT was like a lot of learning in in a week and 181 00:10:19,520 --> 00:10:22,080 that's crazy learning and applying. 182 00:10:22,080 --> 00:10:24,520 And the other end of the day, we also won't. 183 00:10:24,600 --> 00:10:27,320 It was really, really nice feeling as well. 184 00:10:27,840 --> 00:10:31,320 So also encouraged me, OK, I should like to spend more time 185 00:10:31,480 --> 00:10:34,400 focusing and looking for different ideas. 186 00:10:34,960 --> 00:10:39,880 And the main important thing I also mentioned is this is not 187 00:10:40,600 --> 00:10:43,640 ending when you won't or finish the process. 188 00:10:44,000 --> 00:10:46,640 It's just a start. Like the company has started a 189 00:10:46,640 --> 00:10:52,560 huge team over the idea and many other people started to develop 190 00:10:52,560 --> 00:10:54,640 the same thing, but with the standard approach. 191 00:10:54,640 --> 00:10:57,520 Because the Hackerton code is not reusable, it's right only 192 00:10:57,520 --> 00:11:01,920 again. And they started to form a team 193 00:11:02,000 --> 00:11:06,960 and develop and now we have AI3 planner already developed and 194 00:11:06,960 --> 00:11:11,040 available in many regions. And I'm super proud that I like 195 00:11:11,040 --> 00:11:15,320 started this fire and now it's like helping a lot of people. 196 00:11:15,960 --> 00:11:20,440 And this is really, really great feeling that I also had with my 197 00:11:20,440 --> 00:11:25,040 small project, like personally, now I can do it in a larger 198 00:11:25,040 --> 00:11:27,360 scale. I'm super happy and really, 199 00:11:27,360 --> 00:11:29,040 really nice. That's really amazing. 200 00:11:29,400 --> 00:11:32,840 I mean only the fact of booking has thousands of engineers and 201 00:11:32,840 --> 00:11:36,880 if they all get twice a year time to execute on projects like 202 00:11:36,880 --> 00:11:39,840 this and the really good ideas get follow through and actually 203 00:11:39,840 --> 00:11:42,520 make it in the product, that is such great fulfilment. 204 00:11:42,520 --> 00:11:45,080 It's incredible. Exactly, I last time I can 205 00:11:45,080 --> 00:11:49,240 mention it was it was like around 900 people involved. 206 00:11:49,320 --> 00:11:51,600 Yeah, in the hackathon. And that was a lot. 207 00:11:51,600 --> 00:11:55,560 If you walk into the office, see the why people like drawing 208 00:11:55,560 --> 00:11:57,960 something on the whiteboard, like discussing. 209 00:11:58,400 --> 00:12:02,480 And this is not only about like one role in to hackathon. 210 00:12:02,480 --> 00:12:04,880 You can easily switch roles and you can experiment. 211 00:12:04,880 --> 00:12:07,320 For example, if you like to be engineering manager and you are 212 00:12:07,320 --> 00:12:11,240 engineer in the hackathon there there's opportunity to go to the 213 00:12:11,240 --> 00:12:14,680 role and practice and see if you are really good at it. 214 00:12:14,880 --> 00:12:17,400 Yeah, or data engineering, any other role, anything that's not 215 00:12:17,400 --> 00:12:20,240 what you're doing. This is a really good moment to 216 00:12:20,240 --> 00:12:24,800 go to the like different role and see if you like it or not. 217 00:12:24,800 --> 00:12:27,920 So this is really also a really good point for the hackathon. 218 00:12:27,920 --> 00:12:30,320 So it's not just engineering that participates. 219 00:12:30,400 --> 00:12:35,040 Exactly 1 of the project I can mention that was like around we 220 00:12:35,040 --> 00:12:37,640 were around 12 people in the team. 221 00:12:38,080 --> 00:12:40,080 That was a really huge team for that project. 222 00:12:40,520 --> 00:12:45,920 And we had SEO engineer, data analysis, MLPM, EM, all 223 00:12:45,920 --> 00:12:52,080 different roles even copywriter. First I was like thinking, is 224 00:12:52,080 --> 00:12:54,360 that even needed for like what? Are they going to do? 225 00:12:54,520 --> 00:12:57,080 What they're going to do. But at the end of the day, I've 226 00:12:57,080 --> 00:13:01,960 seen each of them has a crucial like point about everything we 227 00:13:01,960 --> 00:13:04,920 do. And we, we have really nice, 228 00:13:04,920 --> 00:13:09,840 like, copywriter giving us right wording for the advertisement, 229 00:13:09,840 --> 00:13:12,600 for the landing pages, for whatever we create. 230 00:13:12,920 --> 00:13:16,440 And the quality at last was like killing quality. 231 00:13:16,800 --> 00:13:19,960 So it was like no brainer for the team to say, OK, you are the 232 00:13:19,960 --> 00:13:20,680 winner. Yeah. 233 00:13:21,000 --> 00:13:24,600 And that that time was I realized what is the value of 234 00:13:24,600 --> 00:13:28,680 these like rows and I learned a lot about different rows, what 235 00:13:28,680 --> 00:13:31,720 is like their mindset. I didn't know about SEO, how 236 00:13:31,720 --> 00:13:33,720 they're going through these steps, how did they find 237 00:13:33,720 --> 00:13:36,360 keywords and other, other stuff around it. 238 00:13:36,360 --> 00:13:40,120 But this is a great learning if you are, if you're not really 239 00:13:40,120 --> 00:13:42,720 happy with what you're doing, This is also a good moment to 240 00:13:42,720 --> 00:13:45,600 see other opportunities, what there are people doing. 241 00:13:45,800 --> 00:13:48,840 If you are interested in you can jump into the different role and 242 00:13:48,840 --> 00:13:53,240 you can transition also as well. Yeah, I really like how you make 243 00:13:53,240 --> 00:13:56,280 the most out of the opportunity, not just from an idea sense, but 244 00:13:56,280 --> 00:13:58,880 from a learning perspective. Learning how to collaborate with 245 00:13:58,880 --> 00:14:01,280 people that you otherwise will never speak kind of in your 246 00:14:01,280 --> 00:14:05,040 day-to-day and seeing and kind of gaining empathy for their 247 00:14:05,040 --> 00:14:07,520 perspective and what they're greater and see how that fits in 248 00:14:07,520 --> 00:14:10,560 a product. I only see upsides basically. 249 00:14:10,920 --> 00:14:13,520 I also participated in a hackathon, Xavier. 250 00:14:13,520 --> 00:14:17,400 We get a one day every other month, which is great, but it's 251 00:14:17,400 --> 00:14:19,200 only one day. And then usually you do like a 252 00:14:19,200 --> 00:14:21,480 proof of concept, you try and do as much in one day. 253 00:14:21,960 --> 00:14:24,720 And then my previous experience at a previous client, it was 254 00:14:24,720 --> 00:14:28,080 like a 2 1/2 day thing. We formed a group and I was 255 00:14:28,080 --> 00:14:30,440 like, OK, this is going to be fun because I never get more 256 00:14:30,440 --> 00:14:32,520 than one day. So we can actually do a lot more 257 00:14:32,520 --> 00:14:34,520 now. And I had this one person who's 258 00:14:34,520 --> 00:14:37,000 team captain, he was like, I've done many hackathons, I've done 259 00:14:37,000 --> 00:14:40,000 like 40. OK, we're going to decide idea 260 00:14:40,000 --> 00:14:41,960 we're going to have he had things time framed. 261 00:14:41,960 --> 00:14:45,160 He really was the one that kept everything on track, make sure 262 00:14:45,160 --> 00:14:48,760 people made good trade-offs, were practical, also delivered 263 00:14:48,760 --> 00:14:51,840 and executed and didn't just overthink or overanalog. 264 00:14:52,760 --> 00:14:55,000 And I think that is really what I needed. 265 00:14:55,960 --> 00:14:59,600 Exactly, so I wasn't that person, but I tend to be that 266 00:14:59,600 --> 00:15:01,960 person. So now I am the person that you 267 00:15:01,960 --> 00:15:04,000 mentioned. So I'm when I the first day, I 268 00:15:04,000 --> 00:15:06,880 always like gather all team. We need to have a strategy. 269 00:15:06,880 --> 00:15:09,640 We need to look into the people that we are like presenting to. 270 00:15:09,920 --> 00:15:13,640 So even list of people that are like in the jury, what is their 271 00:15:13,640 --> 00:15:15,480 mindset? What we should target 272 00:15:15,480 --> 00:15:17,280 everything. So this is like selling the real 273 00:15:17,280 --> 00:15:19,600 product out there. So if you want to sell your 274 00:15:19,600 --> 00:15:22,760 product, you need to look at the customer who is going to like 275 00:15:22,760 --> 00:15:26,560 assess my product. You need to like or like 276 00:15:27,600 --> 00:15:31,400 engineering the process and not just OK, my idea is good, I 277 00:15:31,400 --> 00:15:35,240 develop it. You need to do it like like to 278 00:15:35,240 --> 00:15:38,640 the point and develop something and especially for the demo, 279 00:15:39,160 --> 00:15:43,480 like demo is a really real crucial part of the hackathon 280 00:15:43,800 --> 00:15:46,560 and you need to nail it. And she's not like you have a 281 00:15:46,560 --> 00:15:50,280 great product, sit it there and just your presentation is tool 282 00:15:50,280 --> 00:15:52,640 page and you cannot represent what you've done. 283 00:15:52,640 --> 00:15:55,840 What is the like crucial value behind it For one of the 284 00:15:55,840 --> 00:15:59,480 hackathon, we even created like a production ready videos like 285 00:15:59,960 --> 00:16:02,960 even we interviewed with some people out outside of the 286 00:16:02,960 --> 00:16:06,800 company. So if we like attached a lot of 287 00:16:06,800 --> 00:16:10,200 like content and to make it tangible and also we look into 288 00:16:10,200 --> 00:16:13,240 the feeling of people when you create something, you need to 289 00:16:13,240 --> 00:16:16,200 also touch their feeling. It's not like, OK, my product 290 00:16:16,200 --> 00:16:18,800 does this good. You need to touch the feeling. 291 00:16:18,800 --> 00:16:21,920 So you need to go through the all aspect of a real company 292 00:16:21,920 --> 00:16:25,120 that when you're selling a small start up, this is exactly the 293 00:16:25,120 --> 00:16:27,200 same as selling 1 hackathon project. 294 00:16:27,560 --> 00:16:31,080 You need to go through the same steps and just think like that. 295 00:16:31,080 --> 00:16:34,400 Then it changes everything and all people are getting energy. 296 00:16:34,400 --> 00:16:38,120 OK, we have it like mindset now we can go to this and then the 297 00:16:38,120 --> 00:16:42,360 brainstorming coming and people like sharing their expertise. 298 00:16:42,360 --> 00:16:45,160 We had some people OK, I have a a really good expertise in 299 00:16:45,160 --> 00:16:47,440 narration. I can narrow like create a high 300 00:16:47,440 --> 00:16:49,320 quality narration for the videos. 301 00:16:49,320 --> 00:16:53,560 So what we we created at the end of the day even was amazed of 302 00:16:53,560 --> 00:16:56,680 the quality because so OK, I am really good at editing video. 303 00:16:57,040 --> 00:17:01,520 So they were like like PMEMS but they had some expertise. 304 00:17:01,600 --> 00:17:05,440 They can just turn it into the project and make really the most 305 00:17:05,440 --> 00:17:07,839 out of the project. I think that's gold, like 306 00:17:07,839 --> 00:17:11,440 treating your hackathon project like a mini startup, seeing what 307 00:17:11,440 --> 00:17:13,960 you have in people and leveraging whatever creativity 308 00:17:13,960 --> 00:17:17,040 they have to execute and tell a story and not just create 309 00:17:17,040 --> 00:17:20,520 software, but also sell it to whomever is judging you or 310 00:17:20,520 --> 00:17:22,640 whomever the audience is. Exactly. 311 00:17:22,640 --> 00:17:25,400 That's fantastic. That makes it a lot more like I 312 00:17:25,400 --> 00:17:28,600 get super excited hearing you talk about that environment. 313 00:17:29,200 --> 00:17:31,720 It's really cool that you get time from work to do that. 314 00:17:31,720 --> 00:17:34,240 Have you also seen that outside of company settings like 315 00:17:34,240 --> 00:17:37,280 hackathons that are organized by communities that have the same 316 00:17:37,280 --> 00:17:40,440 vigor? I can say 1 is still is still 317 00:17:40,440 --> 00:17:44,200 inside but it's like we have for the graduates. 318 00:17:44,560 --> 00:17:48,880 So we have same setup for graduate, but it is not one 319 00:17:48,880 --> 00:17:52,640 week, it's 4 weeks. So it is even you have even more 320 00:17:52,640 --> 00:17:57,280 time and that is super amazing because I'm a mentor of those 321 00:17:57,600 --> 00:17:59,600 program. I'm aware of all the steps that 322 00:17:59,600 --> 00:18:02,640 happens. So past year it was like 4. 323 00:18:02,640 --> 00:18:05,840 You have 4 weeks, you join as a graduate, you don't know 324 00:18:05,840 --> 00:18:08,760 anything about like working at the company. 325 00:18:08,760 --> 00:18:10,920 You just jump out of the university. 326 00:18:11,600 --> 00:18:15,200 We want you to know everything after months. 327 00:18:15,400 --> 00:18:18,680 So this is a hackatone you when you join to the first day, you 328 00:18:18,720 --> 00:18:20,960 enter to one of the idea. So it's like joining to 329 00:18:20,960 --> 00:18:23,680 hackatone. You form a team and start 330 00:18:23,680 --> 00:18:27,760 working on the on your project. And each week we do, we'll have 331 00:18:27,760 --> 00:18:29,880 some specific goals. For example, first week you 332 00:18:29,880 --> 00:18:32,920 learn agile, you learn how to work and collaborate with the 333 00:18:32,920 --> 00:18:35,960 teams. Second week you do front end or 334 00:18:35,960 --> 00:18:39,360 back end and 3rd, 3rd week you do front end. 335 00:18:39,960 --> 00:18:43,200 And last week you just form it, create demo and present it. 336 00:18:43,600 --> 00:18:45,840 So these like step by step process. 337 00:18:46,360 --> 00:18:49,160 It's like a month that you have, but it's super intense. 338 00:18:49,600 --> 00:18:51,240 You need to learn a lot of things. 339 00:18:51,240 --> 00:18:53,960 Yeah, imagine you don't know Java, you need to learn it and 340 00:18:53,960 --> 00:18:56,880 you need to do it as well. So with the other help of AI, 341 00:18:56,880 --> 00:19:01,280 it's a bit faster and easier. But in something we do not even 342 00:19:01,280 --> 00:19:05,880 allow them to do, do AI to to try and struggle and know the 343 00:19:05,880 --> 00:19:09,160 value of the stuff, understand and know the technology. 344 00:19:09,160 --> 00:19:11,040 What is the right way of doing that? 345 00:19:11,400 --> 00:19:15,080 Then later this tool can enable you to do that like faster and 346 00:19:15,080 --> 00:19:18,000 so on. But that was really amazing 347 00:19:18,000 --> 00:19:23,160 opportunity for people. And I was mentor of 1 mentor of 348 00:19:23,160 --> 00:19:25,040 one team. I've seen the first day they 349 00:19:25,040 --> 00:19:30,440 struggle even to speak together, but last day of the week, after 350 00:19:30,440 --> 00:19:33,520 like learning sessions and process that we had, they were 351 00:19:33,520 --> 00:19:37,120 really close to each other. They like challenge each other, 352 00:19:37,120 --> 00:19:40,400 they discuss really critical stuff with each other and that 353 00:19:40,440 --> 00:19:42,600 this is exactly what we want as a company. 354 00:19:43,000 --> 00:19:46,840 After a month, if you join to a team, you have almost no 355 00:19:46,840 --> 00:19:48,720 onboarding. You already know all technology 356 00:19:48,720 --> 00:19:53,080 at the company, you already enable yourself using everything 357 00:19:53,080 --> 00:19:55,440 that you probably use later in your team. 358 00:19:56,120 --> 00:19:58,920 And this is going to reduce the onboarding process. 359 00:19:58,920 --> 00:20:02,920 And the feedback we got from the engineer managers was amazing. 360 00:20:02,920 --> 00:20:06,080 They were really amazed with people that Giant just joined 361 00:20:06,080 --> 00:20:09,200 from the university with that level of understanding of the 362 00:20:09,200 --> 00:20:12,440 teamwork and like collaboration with the other people. 363 00:20:13,040 --> 00:20:16,840 And that was one of the example I can say some sort of external 364 00:20:17,480 --> 00:20:20,200 which was really nice. Yeah, interesting. 365 00:20:20,640 --> 00:20:22,680 This for me is fascinating because I've also seen the 366 00:20:22,680 --> 00:20:25,760 industry trend of we don't need junior engineers anymore. 367 00:20:25,760 --> 00:20:27,800 We have AI that does that, even automate it. 368 00:20:27,800 --> 00:20:30,440 And then our seniors, they review PRS and stuff like that. 369 00:20:30,960 --> 00:20:33,840 But the fact that you have junior programs and graduate 370 00:20:33,840 --> 00:20:37,160 programs and people come in, they execute for four weeks and 371 00:20:37,160 --> 00:20:39,800 then they have real praise from engineering managers on that 372 00:20:39,800 --> 00:20:43,280 level is fantastic. Exactly, I'm completely against 373 00:20:43,280 --> 00:20:47,600 this idea because what I've seen as a result, like I was working 374 00:20:47,600 --> 00:20:51,400 directly with them, I was like mentoring them daily. 375 00:20:51,800 --> 00:20:56,080 What I've seen they are really great and we need it like we 376 00:20:56,080 --> 00:21:00,040 need juniors because they, they come from like university, they 377 00:21:00,040 --> 00:21:02,920 have fresh mind. We even had juniors that coming 378 00:21:02,920 --> 00:21:05,920 from university and they had like measured in LLM. 379 00:21:05,920 --> 00:21:08,560 So they have like top knowledge about LLM. 380 00:21:09,040 --> 00:21:12,920 We want them because if they we move them to to the right team, 381 00:21:12,920 --> 00:21:15,600 they can like perform 10 times better than other people. 382 00:21:15,600 --> 00:21:18,480 They do not know, do not know anything about LLM, for example. 383 00:21:19,160 --> 00:21:22,640 And throughout this process we could like enable them in a 384 00:21:22,640 --> 00:21:25,200 month. So I don't think it's still like 385 00:21:25,200 --> 00:21:27,840 they're not valuable. They're super valuable and they 386 00:21:27,840 --> 00:21:30,560 could like perform really well. So I know a lot of them. 387 00:21:30,560 --> 00:21:34,880 I'm now friend with them. I wasn't their mental, but 388 00:21:34,880 --> 00:21:36,600 they're really performed really well. 389 00:21:36,600 --> 00:21:39,560 And all people even like engineers are amazed. 390 00:21:39,560 --> 00:21:42,160 Is that really a graduate that joined our team? 391 00:21:42,160 --> 00:21:45,600 They're really good and they can perform really well and with the 392 00:21:45,600 --> 00:21:50,000 help of AI, if they're using it the right way and first go 393 00:21:50,000 --> 00:21:53,120 through the understanding of the factors, not jumping into the 394 00:21:53,120 --> 00:21:57,240 conclusion, they always can have a proper result as well in 395 00:21:57,240 --> 00:21:59,920 consultation with the other seniors around them. 396 00:22:00,080 --> 00:22:03,120 Yeah, sounds like a lot of fun. Like if I could go back in time, 397 00:22:03,120 --> 00:22:05,000 I would have loved to join a graduate program. 398 00:22:05,000 --> 00:22:06,040 Like this? Definitely. 399 00:22:06,040 --> 00:22:08,400 Sounds like so much fun. Exactly how many people are 400 00:22:08,400 --> 00:22:11,000 usually in the graduate program that do this for a week? 401 00:22:11,920 --> 00:22:16,440 Last time was like 770. 70 Yeah, that's incredible, yeah. 402 00:22:16,720 --> 00:22:20,400 Really huge amount of people and they're really crazy with the 403 00:22:20,400 --> 00:22:23,840 idea they have a fresh mind. They come up with their own 404 00:22:23,840 --> 00:22:26,040 ideas. There's some sort of like ground 405 00:22:26,040 --> 00:22:29,520 ideas coming from the product managers because you don't want 406 00:22:29,520 --> 00:22:35,080 to just like put their like what they do just they want to apply 407 00:22:35,080 --> 00:22:38,400 something into it and they want to use those projects and they 408 00:22:38,400 --> 00:22:41,920 want we want to treat it as a like production product to make 409 00:22:41,920 --> 00:22:45,440 them also like doing something serious, not just OK, I'm going 410 00:22:45,440 --> 00:22:47,240 to do it and they're going to just delete it. 411 00:22:47,240 --> 00:22:49,240 At the end of the day, this is not the case. 412 00:22:49,240 --> 00:22:51,760 You want to make them more serious like this is something 413 00:22:51,760 --> 00:22:53,240 that we're going to use it later. 414 00:22:53,240 --> 00:22:56,240 So they are mostly OK. The algorithm should be like 415 00:22:56,240 --> 00:22:58,400 this. And so the discussion changes 416 00:22:58,400 --> 00:23:02,360 when you note the thing that your looking is seriously going 417 00:23:02,360 --> 00:23:06,440 to be used later. But they have like capability to 418 00:23:06,440 --> 00:23:09,360 change it to to what extent they want because they can add a new 419 00:23:09,360 --> 00:23:12,360 feature, they can change the direction, but the main goal is 420 00:23:12,360 --> 00:23:14,720 like already defined. You want to have something with 421 00:23:14,760 --> 00:23:16,920 XYZ, but you can do whatever you want. 422 00:23:17,280 --> 00:23:19,880 So there's no limit into the functionality and features. 423 00:23:19,880 --> 00:23:22,480 They can just come up with the crazy ideas if they can 424 00:23:22,480 --> 00:23:25,760 implement it. I feel like that is very smart 425 00:23:26,280 --> 00:23:27,600 and I don't know if I've heard that before. 426 00:23:27,600 --> 00:23:31,040 Like if I think of hackathons, it's usually you have a theme 427 00:23:31,040 --> 00:23:33,960 and you can do whatever you want within that theme, but if you 428 00:23:33,960 --> 00:23:36,120 are creating something and there's a chance that it will be 429 00:23:36,120 --> 00:23:38,400 used in the eventual product, you act differently. 430 00:23:38,880 --> 00:23:41,400 I think that's very smart, especially for a product company 431 00:23:41,680 --> 00:23:43,720 to say, OK, these are our products and product managers 432 00:23:43,720 --> 00:23:45,760 have thought of these ideas. You can bring in your own or you 433 00:23:45,760 --> 00:23:49,120 can participate in this and then it might actually contribute and 434 00:23:49,120 --> 00:23:50,400 benefit. Usually when you're talking 435 00:23:50,400 --> 00:23:53,320 about hackathons, if you talk to the wrong people, they see it as 436 00:23:53,320 --> 00:23:55,920 a cost, right? What do you mean 3 days? 437 00:23:55,920 --> 00:23:57,760 People don't work and they just play around what? 438 00:23:57,760 --> 00:24:00,320 What is the benefit? Basically, you can say something 439 00:24:00,320 --> 00:24:03,160 vague with regards to knowledge sharing and learning, stuff like 440 00:24:03,160 --> 00:24:05,840 that, but if in the end there's something tangible that comes 441 00:24:05,840 --> 00:24:08,240 out of it that we can actually put to production and has 442 00:24:08,240 --> 00:24:10,080 business impact. Exactly. 443 00:24:10,080 --> 00:24:11,720 And all of a sudden you're like, oh, that's interesting. 444 00:24:11,920 --> 00:24:15,000 Yeah, what what like what the booking does is like goes from 445 00:24:15,000 --> 00:24:17,560 the other side, it's sending a message to all PMS. 446 00:24:17,560 --> 00:24:20,760 Do you have anything you need and you do not have capability 447 00:24:20,760 --> 00:24:24,360 of doing that. So they're listing those up and 448 00:24:24,360 --> 00:24:25,960 then they're looking into the items. 449 00:24:25,960 --> 00:24:29,080 OK, which one is more suitable for these people that are coming 450 00:24:29,440 --> 00:24:33,560 because the project itself built with the like mentorship. 451 00:24:33,560 --> 00:24:35,640 It's not like they're developing by themselves. 452 00:24:35,640 --> 00:24:38,640 There's a mentor always looking into the all process. 453 00:24:38,960 --> 00:24:42,320 So the quality, the the output is really high, the quality of 454 00:24:42,320 --> 00:24:46,560 the output and they cannot be used directly, but they are 455 00:24:46,560 --> 00:24:50,840 really ground like floor for developing a real like 456 00:24:51,200 --> 00:24:55,280 production ready product out. So we're using high tech, the 457 00:24:55,320 --> 00:24:58,960 high technology and like cloud and everything that you're using 458 00:24:58,960 --> 00:25:02,560 daily because you want them to experience what they're they're 459 00:25:02,560 --> 00:25:05,560 trying to do later. So they're like small amount of 460 00:25:05,560 --> 00:25:09,400 experience before jumping into real work and you experience 461 00:25:09,400 --> 00:25:12,040 everything. Like we try to allow them to 462 00:25:12,040 --> 00:25:16,160 experience all aspects because in large, large organizations, 463 00:25:16,160 --> 00:25:19,080 you may go to a team that you have like working only a 464 00:25:19,080 --> 00:25:23,040 specific on a specific feature. So you do not allow to work on 465 00:25:23,040 --> 00:25:25,560 everything. But we want them to already 466 00:25:25,560 --> 00:25:28,600 experience everything. Different aspects of the like 467 00:25:28,600 --> 00:25:32,480 software engineering and some of them even rotate into different 468 00:25:32,480 --> 00:25:35,800 teams, like 6-1 to the team one and six one to team 2. 469 00:25:35,800 --> 00:25:39,200 Because if we want to want them to have exposure to different 470 00:25:39,240 --> 00:25:42,320 aspect of the product because it's too big and at least a lot 471 00:25:42,320 --> 00:25:46,480 of stuff to learn. And this is like a really ground 472 00:25:46,480 --> 00:25:49,200 learning process. And then they jump to the like 473 00:25:49,280 --> 00:25:53,000 2-6 months process that they, they will rotate into different 474 00:25:53,000 --> 00:25:56,200 teams and learn more and more. And after that they they get 475 00:25:56,840 --> 00:26:01,280 automatic promotion and then they'll go to the real life team 476 00:26:01,280 --> 00:26:05,440 and stop like stay in one team. And this process I think is 477 00:26:05,440 --> 00:26:08,960 really a smart design, has a really smart design that you 478 00:26:09,320 --> 00:26:12,760 like stop drum intense when the first day that you work at the 479 00:26:12,760 --> 00:26:15,480 team, you feel like, OK, this is a relief for me. 480 00:26:15,800 --> 00:26:19,480 So, but they learned a lot. So the process, they're learning 481 00:26:19,480 --> 00:26:23,480 a lot and they learn not only for hard skill, but we had huge 482 00:26:23,480 --> 00:26:27,120 focus on the soft skill and agile and like teamwork. 483 00:26:27,600 --> 00:26:30,080 So we have so many more programming, like per 484 00:26:30,080 --> 00:26:33,120 programming, so, so many things that to connect them with with 485 00:26:33,120 --> 00:26:36,880 each other and know how to work with each other in real, real 486 00:26:36,880 --> 00:26:37,680 life. Yeah. 487 00:26:38,280 --> 00:26:40,160 And that's really important. Yeah, interesting. 488 00:26:40,720 --> 00:26:42,800 And you've been in many hackathon teams and you've 489 00:26:42,800 --> 00:26:45,640 mentored graduates that also go through this process. 490 00:26:46,120 --> 00:26:48,120 What are some of the things that people struggle with with 491 00:26:48,120 --> 00:26:51,200 regards to executing and delivering and kind of reaching 492 00:26:51,200 --> 00:26:52,520 the goal that they set for themselves? 493 00:26:52,520 --> 00:26:55,200 Is it planning? Is it actually teamwork or 494 00:26:55,200 --> 00:26:57,040 execution? What is usually a challenge? 495 00:26:57,520 --> 00:27:01,280 One of the challenge I've seen is when, when you open like the 496 00:27:01,360 --> 00:27:06,360 capability of the new features, they may like aim for the really 497 00:27:06,360 --> 00:27:09,680 high and they cannot reach it. So they, they do not know how to 498 00:27:09,680 --> 00:27:13,440 easily break down the challenges to the small challenges and go 499 00:27:13,440 --> 00:27:16,600 step by step. Sometimes they aim really high 500 00:27:16,600 --> 00:27:18,880 and they cannot even reach it. So they're disappointed, they 501 00:27:18,880 --> 00:27:22,960 change direction and so on. So as a mentor, what I do is 502 00:27:22,960 --> 00:27:27,640 always sit first, sit and look. They're struggling. 503 00:27:27,640 --> 00:27:32,280 So I do not like jump in. You need to wait and see or how 504 00:27:32,280 --> 00:27:34,680 they're struggling. What is their then you need to 505 00:27:34,680 --> 00:27:36,960 guide them, not even directly give them the answer. 506 00:27:36,960 --> 00:27:39,160 And what if we do this And what if we do that? 507 00:27:39,160 --> 00:27:43,320 So this way you help them in order to think that way and go 508 00:27:43,320 --> 00:27:46,000 ahead and figure out the solution by themselves. 509 00:27:46,480 --> 00:27:50,560 So I, I, when I just guide them through the, these challenges, 510 00:27:50,560 --> 00:27:52,520 they end. OK, We should do it like small 511 00:27:52,520 --> 00:27:55,280 by step by step. We should break it down if and 512 00:27:55,280 --> 00:27:59,040 then they managed to do it. And that was one of the thing 513 00:27:59,040 --> 00:28:00,960 I've seen. And the other thing was more 514 00:28:00,960 --> 00:28:04,880 about the collaboration, the the issue that he had, they had they 515 00:28:04,880 --> 00:28:07,560 didn't know how to collaborate well with each other. 516 00:28:07,600 --> 00:28:12,160 Everyone was really great solo, but they couldn't work together. 517 00:28:12,160 --> 00:28:15,720 They couldn't enable each other and that was also something that 518 00:28:15,720 --> 00:28:19,200 we were working a lot on on the process to enable them. 519 00:28:19,200 --> 00:28:22,400 We did like more programming is one of the thing that starts and 520 00:28:22,400 --> 00:28:25,920 breaks the ice between them. Yeah, because one of them you 521 00:28:25,920 --> 00:28:27,720 need to see it and the other need to guide. 522 00:28:27,720 --> 00:28:30,800 And so did this speaking happens technical speaking question 523 00:28:30,800 --> 00:28:32,760 answer. So the desk, this is really nice 524 00:28:32,760 --> 00:28:36,400 start world for the for the enabling them for the 525 00:28:36,400 --> 00:28:37,440 collaboration. Yeah. 526 00:28:37,680 --> 00:28:39,640 Is that the same for experienced engineers? 527 00:28:39,720 --> 00:28:43,360 I, I can imagine that if I've never done a hackathon and I go 528 00:28:43,360 --> 00:28:46,920 into a team, I do kind of know how to break stuff down. 529 00:28:46,920 --> 00:28:49,200 I might fall into kind of analysis paralysis. 530 00:28:49,480 --> 00:28:52,360 Collaboration might always be challenging depending on if 531 00:28:52,360 --> 00:28:54,160 you've more programmed, if you know how to break those 532 00:28:54,160 --> 00:28:56,480 barriers. But it's are those the same 533 00:28:56,480 --> 00:28:58,680 challenges you've seen in existing hackathon teams? 534 00:28:58,960 --> 00:29:02,400 I would say kind of kind of the same challenges, but each 535 00:29:02,520 --> 00:29:05,520 hackathon team needs someone to guide and lead. 536 00:29:05,520 --> 00:29:08,520 So that's why I always encourage EMS to join the hackathons 537 00:29:09,000 --> 00:29:11,400 because they already have the capability and knowledge of 538 00:29:11,400 --> 00:29:14,200 managing teams. And if they join hackathons, 539 00:29:14,200 --> 00:29:16,480 most of the time they're not joining because, OK, it's too 540 00:29:16,480 --> 00:29:17,840 technical, I don't want to join it. 541 00:29:17,840 --> 00:29:22,520 But we want EMS in hackathons and if they join, they can 542 00:29:22,520 --> 00:29:25,760 easily guide and help and do whatever they do in real life, 543 00:29:26,960 --> 00:29:30,040 but a bit more fast in a bit more fast peace environment. 544 00:29:30,360 --> 00:29:33,800 They can guide people, they can like lead people in order to 545 00:29:34,120 --> 00:29:36,280 each team. A successful team hackathon 546 00:29:36,280 --> 00:29:39,840 needs someone to do that, even if it's not like he or she is 547 00:29:39,840 --> 00:29:43,160 not a an engineer manager. It needs someone to manage 548 00:29:43,160 --> 00:29:47,120 everything and blocks ideas because so many ideas going to 549 00:29:47,160 --> 00:29:52,560 come up and we need to like have a some limit for our focus on 550 00:29:52,560 --> 00:29:57,120 what we're going to deliver and then expand, expand like when 551 00:29:57,240 --> 00:30:01,600 times like times time allows. So that's really important. 552 00:30:01,600 --> 00:30:04,120 Like member of each successful hackathon teams. 553 00:30:04,680 --> 00:30:08,320 If all like engineers sitting together just coding even the 554 00:30:08,320 --> 00:30:10,760 quality, if the quality of the product is super great. 555 00:30:10,760 --> 00:30:15,360 They do do like ground breaking project is there, it's not going 556 00:30:15,360 --> 00:30:16,560 to be recognized. Yeah. 557 00:30:16,800 --> 00:30:21,080 And they need someone to focus on the goal and like why we 558 00:30:21,080 --> 00:30:25,200 doing that always should be this question and then getting to the 559 00:30:25,200 --> 00:30:29,000 result then adding up some like fancy feature over that. 560 00:30:30,200 --> 00:30:33,080 I receive a lot of questions for people that are slightly more 561 00:30:33,080 --> 00:30:36,680 early in career with regards to how can I distinguish myself. 562 00:30:36,680 --> 00:30:39,080 Nowadays in this tech field, people are getting laid off, 563 00:30:39,400 --> 00:30:41,680 there's very good expertise out in the market and I'm applying 564 00:30:41,680 --> 00:30:43,520 for jobs, but how can I distinguish myself? 565 00:30:44,080 --> 00:30:46,880 I feel like there's a lot of benefits that you can get from 566 00:30:46,880 --> 00:30:49,080 hackathon, but I'm very curious to hear your perspective. 567 00:30:49,480 --> 00:30:52,880 What would you say the biggest benefits are of participating in 568 00:30:52,880 --> 00:30:57,600 in hackathons as engineers? For me, for for me myself, it 569 00:30:57,600 --> 00:31:02,360 was like the most benefit I get was not like the prize was not 570 00:31:02,360 --> 00:31:04,840 the, the like other stuff around. 571 00:31:04,840 --> 00:31:06,640 It was the learning process I had. 572 00:31:07,000 --> 00:31:11,040 So for the softer skill, you have huge amount of opportunity 573 00:31:11,040 --> 00:31:12,640 to to learn because you have one week. 574 00:31:12,640 --> 00:31:15,160 You need to connect with people, you need to collaborate with 575 00:31:15,160 --> 00:31:18,280 people, you need to get a like something out with those people. 576 00:31:18,280 --> 00:31:21,800 So this is really challenging. We normally when you like on 577 00:31:21,800 --> 00:31:24,520 boarding to a team, it takes a month or something to get the 578 00:31:24,520 --> 00:31:26,880 same like connection with the people. 579 00:31:27,280 --> 00:31:31,040 But because you, you have this urge, you don't have time, it 580 00:31:31,040 --> 00:31:34,600 allows you to jump and it allows to over perform and doing that. 581 00:31:34,600 --> 00:31:37,480 So, so it was like for me, it was really nice. 582 00:31:38,120 --> 00:31:40,880 It's not even possible to connect for those people. 583 00:31:41,400 --> 00:31:46,120 And now they're like all teams. I was working now they're my 584 00:31:46,120 --> 00:31:49,120 close friends. So it was like the time we were 585 00:31:49,120 --> 00:31:52,880 really in connection that we changed to be a friend because 586 00:31:52,960 --> 00:31:57,160 even like more than like my connection is more with like my 587 00:31:57,160 --> 00:32:01,200 day-to-day, like teammates because time is really intense. 588 00:32:01,360 --> 00:32:04,880 You need to like focus. You need to put your whole 589 00:32:04,880 --> 00:32:08,480 attention into the project and then you need to collaborate 590 00:32:08,480 --> 00:32:11,600 well with all people. And not only for that, for the 591 00:32:11,600 --> 00:32:14,760 technical aspect is amazing. For example, for prompt 592 00:32:14,760 --> 00:32:17,600 engineering, that was like my enter point to the prompt 593 00:32:17,600 --> 00:32:21,760 engineering understanding about that, like getting my hand dirty 594 00:32:21,760 --> 00:32:24,880 with with AILM. So maybe your day-to-day work, 595 00:32:24,880 --> 00:32:26,720 you don't have capability to do that. 596 00:32:26,720 --> 00:32:29,880 So you're doing working something completely not related 597 00:32:29,880 --> 00:32:32,480 to that. But hackathon is a great time to 598 00:32:32,480 --> 00:32:36,640 to just go out there. Even if you find a project that 599 00:32:36,640 --> 00:32:39,400 is that they're using some technology that you like, just 600 00:32:39,400 --> 00:32:41,600 jump into it. This is the best moment you can 601 00:32:41,600 --> 00:32:43,520 learn. You can apply the knowledge in 602 00:32:43,520 --> 00:32:48,280 real life project and I'm using it always like like this because 603 00:32:48,320 --> 00:32:52,880 I'm also like into gaming and developing games, but really 604 00:32:52,880 --> 00:32:56,360 hobby games. As I mentioned, I found same 605 00:32:56,360 --> 00:32:58,760 like opportunity in one of the hackathon there like there was a 606 00:32:58,760 --> 00:33:00,520 team, they wanted to develop a game. 607 00:33:00,760 --> 00:33:03,320 I jump into the idea because, OK, this is a really good 608 00:33:03,320 --> 00:33:05,640 moment. I cannot find any opportunity to 609 00:33:05,640 --> 00:33:09,360 develop a game, a booking. So it doesn't have any relation. 610 00:33:10,120 --> 00:33:15,160 So the game itself was was to like enable people to learn. 611 00:33:15,160 --> 00:33:19,400 So we have a an experimentation, famous experimentation tooling 612 00:33:19,400 --> 00:33:21,120 that we developed for a long time. 613 00:33:21,120 --> 00:33:26,000 And because like experimentation is DNA of booking.com. 614 00:33:26,360 --> 00:33:29,320 We have so many AB, TS and so on, But the tool getting really 615 00:33:29,320 --> 00:33:31,800 complex because it developed throughout many years. 616 00:33:31,800 --> 00:33:33,880 And if you want to learn it, it takes a lot of time. 617 00:33:34,320 --> 00:33:39,280 So the idea was to create a game and let people play the game and 618 00:33:39,480 --> 00:33:42,720 you getting to to a blocker and then you need to open 619 00:33:42,720 --> 00:33:44,720 experimentation tool and do some action. 620 00:33:45,240 --> 00:33:49,440 Then you go go to next step and it asks you now unblock. 621 00:33:49,440 --> 00:33:53,280 So you need to you you learn a process without even opening 622 00:33:53,280 --> 00:33:56,360 documentation. It's like a it was like a Star 623 00:33:56,360 --> 00:34:00,560 Wars style game and you need to go to the step by step answering 624 00:34:00,560 --> 00:34:03,200 the question and some sometimes open this link and go to the 625 00:34:03,200 --> 00:34:04,840 experimentation. Now you need to set up the 626 00:34:04,840 --> 00:34:09,159 experimentation and go ahead. And because of the time, the 627 00:34:09,159 --> 00:34:11,920 experimentation always time is important because you need to 628 00:34:11,920 --> 00:34:13,920 wait for a month to get the result. 629 00:34:13,920 --> 00:34:17,960 But we speed that up with create like simulating the date and 630 00:34:17,960 --> 00:34:19,520 data. We inject data to the 631 00:34:19,520 --> 00:34:23,159 experimentation like simulating that like a month past. 632 00:34:23,520 --> 00:34:25,960 Now you need to make decision. Now you need to make a call. 633 00:34:25,960 --> 00:34:27,920 Is that going to be full on or not? 634 00:34:28,239 --> 00:34:31,080 So this is really crucial and important decision for the 635 00:34:31,080 --> 00:34:33,320 company. But we changed it to a game and 636 00:34:33,320 --> 00:34:36,440 you can play thousand time and then you can understand and 637 00:34:36,440 --> 00:34:39,760 learn how to make the call in different situation. 638 00:34:40,040 --> 00:34:43,840 And the benefit goes directly to the company because this is 639 00:34:43,960 --> 00:34:46,199 really crucial. If you call and make it run 640 00:34:46,199 --> 00:34:48,679 call, you're going to get like a lot lost a lot of money. 641 00:34:48,880 --> 00:34:53,239 So we created a game. You do not need to open even a 642 00:34:53,239 --> 00:34:56,120 documentation and go through the learning process super fast. 643 00:34:56,639 --> 00:34:59,000 So it has different levels. You start really simple 644 00:34:59,000 --> 00:35:01,640 experiment getting difficult, difficult and difficult. 645 00:35:02,320 --> 00:35:06,640 And that was only in a week. So it was really amazed. 646 00:35:06,640 --> 00:35:09,560 The team were really happy that we could manage this and the 647 00:35:09,640 --> 00:35:13,240 output was really tangible. Maybe this is not for everyone. 648 00:35:13,360 --> 00:35:16,080 Some people are like video and so on, but there is a lot of 649 00:35:16,080 --> 00:35:19,840 people they want to get hand hands on the project and see 650 00:35:19,840 --> 00:35:23,600 what is what he's doing in real life and make a call and see if 651 00:35:23,600 --> 00:35:26,400 he's not. And it's like like safe call. 652 00:35:26,760 --> 00:35:30,040 So if you make this like wrong decision, you just learn and 653 00:35:30,040 --> 00:35:32,000 your company doesn't lose anything. 654 00:35:32,480 --> 00:35:36,600 And that was really important. This is like one of the thing I 655 00:35:36,600 --> 00:35:39,240 mentioned that you need if you want to learn something, this is 656 00:35:39,240 --> 00:35:41,680 the best moment. You need to look into project 657 00:35:41,800 --> 00:35:45,360 that they have a goal or aim to do something that you want to 658 00:35:45,360 --> 00:35:46,400 learn. Jump into it. 659 00:35:46,400 --> 00:35:49,440 There is no need to having the expertise to jump. 660 00:35:50,000 --> 00:35:53,160 This is the moment you need to go and learn and do it. 661 00:35:53,280 --> 00:35:57,040 Yeah, And I feel like what what would hold people back is 662 00:35:57,040 --> 00:35:59,000 they're afraid, right? How can I do something that I've 663 00:35:59,000 --> 00:36:01,880 never done before? I feel like nowadays that 664 00:36:01,880 --> 00:36:04,160 barrier of entry is lower and lower because you have a lot of 665 00:36:04,160 --> 00:36:06,800 tooling exactly and you are able to educate yourself. 666 00:36:06,800 --> 00:36:09,960 So then it's just confidence. It's way better than before. 667 00:36:10,000 --> 00:36:11,520 Yeah. Yeah, exactly. 668 00:36:12,280 --> 00:36:15,320 So you just have to do it, man. I do hate that answer because 669 00:36:15,480 --> 00:36:18,280 for me it's very hard to start. And then people say, you know 670 00:36:18,280 --> 00:36:20,720 what makes it easier to start? It's just to do it and to do it 671 00:36:20,720 --> 00:36:23,960 over and over and over again. It sounds very simple and still 672 00:36:23,960 --> 00:36:26,080 it's challenging, right? Building something, having 673 00:36:26,080 --> 00:36:27,880 ideas. I mean, even when you and I 674 00:36:27,880 --> 00:36:31,120 talked over lunch just now, you were already talking about ideas 675 00:36:31,120 --> 00:36:34,640 and you see those ideas due to the experience you have building 676 00:36:34,640 --> 00:36:37,760 up on your ideas. I feel like sometimes I have 677 00:36:37,760 --> 00:36:40,800 ideas and the step to actually building those, they do become 678 00:36:40,800 --> 00:36:43,040 smaller. I have fun with vibe coding. 679 00:36:43,240 --> 00:36:45,360 I know start easier, but I don't finish. 680 00:36:45,360 --> 00:36:48,280 For example, I have ideas that I'm like, oh, can I make a game 681 00:36:48,280 --> 00:36:50,240 out of that? Then I get to a point where I 682 00:36:50,240 --> 00:36:53,480 can make a lobby, people can join on the same room, but then 683 00:36:53,480 --> 00:36:55,400 I don't actually continue with creating the game. 684 00:36:55,400 --> 00:36:58,160 I start and do something else which I've never had before. 685 00:36:58,640 --> 00:37:01,360 So the dynamic changes, but I am learning a lot. 686 00:37:01,360 --> 00:37:04,120 I feel like this pressure cooker, both for your soft 687 00:37:04,120 --> 00:37:06,400 skills and for your hard skills, is very valuable. 688 00:37:06,920 --> 00:37:10,760 Exactly the same, but about the idea is like I've created a 689 00:37:10,760 --> 00:37:12,560 project, it's like a personal project. 690 00:37:12,840 --> 00:37:16,440 I wanted to know more about sentiment analysis, like and 691 00:37:16,440 --> 00:37:18,840 like intention and topic detection and so on. 692 00:37:18,920 --> 00:37:22,160 But I wanted to do it in a safe environment for my personal 693 00:37:22,160 --> 00:37:22,840 data. Yeah. 694 00:37:23,080 --> 00:37:26,920 So I've created a Telegram bot and attach it to the friend 695 00:37:26,920 --> 00:37:29,120 group that we have. So there's a group that we have 696 00:37:29,120 --> 00:37:31,400 our friends and this is a safe place. 697 00:37:31,400 --> 00:37:35,320 I can easily attach everything to that and this telegram bot 698 00:37:35,320 --> 00:37:40,320 just collect all messages and send it back to my LLM server 699 00:37:40,320 --> 00:37:43,800 that I have. And what I do, I analyse the 700 00:37:44,040 --> 00:37:48,360 like sentiment, the topics people speaking about is the 701 00:37:48,360 --> 00:37:51,440 trend goes up, goes down. I create a dashboard, fancy 702 00:37:51,440 --> 00:37:53,960 dashboard. But each time I'm looking at it, 703 00:37:54,720 --> 00:37:58,200 because this was a basic idea, I come up with a new idea. 704 00:37:58,200 --> 00:38:00,160 OK, what if I predict the next message? 705 00:38:00,160 --> 00:38:02,760 So I added something to predict in my next message. 706 00:38:03,120 --> 00:38:06,480 So each time the message I get the context of the previous 707 00:38:06,520 --> 00:38:10,080 messages and message they reply and try to predict what is the 708 00:38:10,080 --> 00:38:11,760 next message. Then it compares with the next 709 00:38:11,760 --> 00:38:13,920 message. What is the accuracy of the 710 00:38:13,920 --> 00:38:17,120 prediction? So there are many ideas coming 711 00:38:17,120 --> 00:38:18,680 up. Each time you look into the 712 00:38:18,680 --> 00:38:22,120 previous idea, you need to start the the, the the meaning of 713 00:38:22,320 --> 00:38:25,920 start is to if you have something there, it's like 714 00:38:25,920 --> 00:38:30,000 giving you energy to to add some new brick back over that. 715 00:38:30,240 --> 00:38:33,280 And then you see you ending huge project. 716 00:38:33,560 --> 00:38:36,320 But sometimes you aim really high and you want to, for 717 00:38:36,320 --> 00:38:41,400 example, add something. And this is really too big to be 718 00:38:41,400 --> 00:38:43,560 very feasible and to disappoint you to do it. 719 00:38:43,560 --> 00:38:47,320 So need to go really small, really like much stuff, just go 720 00:38:47,320 --> 00:38:52,160 really small steps, small steps. Then then it encourages you to 721 00:38:52,160 --> 00:38:55,320 do it more and more. So this is for me it was working 722 00:38:55,560 --> 00:38:57,440 like doing really small features. 723 00:38:57,440 --> 00:39:00,440 So each time I didn't develop the whole idea together. 724 00:39:00,440 --> 00:39:03,840 The first version only had like showing the message that was in 725 00:39:03,840 --> 00:39:06,320 the tag of nothing out without LLM. 726 00:39:06,680 --> 00:39:09,120 But the current version, even I can switch between LLM. 727 00:39:09,120 --> 00:39:11,000 I can reanalyze. You can change the prompt. 728 00:39:11,280 --> 00:39:15,360 So each time I added a new feature, I then it it gives me 729 00:39:15,360 --> 00:39:18,440 huge energy. It's adding a new feature to it. 730 00:39:18,760 --> 00:39:20,680 Let's do it. So even in midnight, I would 731 00:39:20,680 --> 00:39:22,560 say, OK, I need to add something to it. 732 00:39:22,560 --> 00:39:24,200 So I'd see what's going to happen. 733 00:39:24,640 --> 00:39:27,160 And this is like also sharing some of that with my friend and 734 00:39:27,200 --> 00:39:29,560 amaze. Oh my God, have you done that? 735 00:39:29,560 --> 00:39:32,680 So it's so nice and that's really nice. 736 00:39:32,680 --> 00:39:36,720 I think this stuff like you can do and the learning process, 737 00:39:36,720 --> 00:39:40,640 like I'm doing the same work at my current job nowadays, but the 738 00:39:40,640 --> 00:39:43,160 scale is huge. This is completely different. 739 00:39:43,160 --> 00:39:47,600 But this learning process helped me really nice and like enabled 740 00:39:47,600 --> 00:39:52,360 me to do it involved in all steps, which was even not like 741 00:39:52,560 --> 00:39:55,320 part of my job description. I could involve. 742 00:39:55,320 --> 00:39:59,440 I could like have a huge and dramatic effect into many like 743 00:39:59,440 --> 00:40:03,320 prompting of the process and any other process related to LLM and 744 00:40:04,120 --> 00:40:09,080 AI related to my real job. But it started all from the 745 00:40:09,080 --> 00:40:11,440 hobby project. It's like my I'm kind of 746 00:40:11,440 --> 00:40:14,880 addicted to this nowadays. I can see that. 747 00:40:14,880 --> 00:40:18,960 I think it's, I think it's very smart to figure out a way to get 748 00:40:18,960 --> 00:40:22,400 really good at executing with different technologies, building 749 00:40:22,400 --> 00:40:25,360 little projects here and there for your own kind of personal 750 00:40:25,360 --> 00:40:27,520 journey. I feel like it's really good in 751 00:40:27,520 --> 00:40:29,480 distinguishing yourself when you're talking about career 752 00:40:29,480 --> 00:40:31,960 perspective. But also like you mentioned, you 753 00:40:31,960 --> 00:40:33,200 can apply it at your current job. 754 00:40:33,200 --> 00:40:36,320 You can actually see other teams that are doing something that 755 00:40:36,320 --> 00:40:38,560 actually you've done in a personal project or you're like, 756 00:40:38,560 --> 00:40:41,440 huh, how would you do that? You execute and then because 757 00:40:41,440 --> 00:40:43,240 you're having so much fun, you move to that team. 758 00:40:43,680 --> 00:40:46,240 I think it's fantastic. Yeah, exactly, exactly. 759 00:40:46,520 --> 00:40:49,040 And that helped me a lot. You know, for example, the same 760 00:40:49,280 --> 00:40:51,320 project I mentioned. I'm doing the same but with the 761 00:40:51,320 --> 00:40:53,840 larger scale and. Booking scale. 762 00:40:53,840 --> 00:40:57,840 In the booking scale, so, so I, I know both aspects. 763 00:40:57,840 --> 00:41:01,600 I know like how how it goes with the small project and how it 764 00:41:01,600 --> 00:41:03,200 goes. What's the challenges of the big 765 00:41:03,200 --> 00:41:05,960 projects. So, and I can apply some of them 766 00:41:05,960 --> 00:41:08,920 to to the other. And even I learned something my 767 00:41:08,920 --> 00:41:10,720 job. OK, let's apply to my personal 768 00:41:10,720 --> 00:41:11,160 project. Yeah. 769 00:41:11,520 --> 00:41:13,400 And improve it. And so we had so, so many 770 00:41:13,400 --> 00:41:15,920 meetings with open AI that we are working directly. 771 00:41:16,440 --> 00:41:19,760 And I learned a lot about how, how you need to like, for 772 00:41:19,760 --> 00:41:22,120 example, use model. You need to solve with a larger 773 00:41:22,120 --> 00:41:25,080 model, then you need to adopt it to the medium model. 774 00:41:25,320 --> 00:41:27,560 And the final project you can use is be the flash, for 775 00:41:27,560 --> 00:41:29,800 example. But you need to fine tune, you 776 00:41:29,800 --> 00:41:32,520 need to like adopt it. Don't start with the like 777 00:41:32,720 --> 00:41:35,280 cheaper model because you're going to get the poor result and 778 00:41:35,280 --> 00:41:38,600 you don't know how to like fine tune the the prompt that you 779 00:41:38,600 --> 00:41:40,960 have. So this learning a really 780 00:41:40,960 --> 00:41:43,760 general learning that you can use in your personal project or 781 00:41:43,760 --> 00:41:48,240 the real life project, but you can experiment it easily because 782 00:41:48,440 --> 00:41:51,400 if you want to experiment in the real world, it's more 783 00:41:51,400 --> 00:41:53,640 challenging. But experimenting in my personal 784 00:41:53,640 --> 00:41:56,720 project related to my friend channel is super simple. 785 00:41:57,240 --> 00:41:59,760 And also I can experiment whatever I want, different 786 00:41:59,760 --> 00:42:03,480 languages, what was like the output for the different models 787 00:42:03,480 --> 00:42:05,800 and everything. And it goes with no risk. 788 00:42:06,560 --> 00:42:09,040 Yeah, I think that's the best. I I love learning about your 789 00:42:09,080 --> 00:42:10,560 projects and your passion for hackathon. 790 00:42:11,360 --> 00:42:12,920 Thank you so much for coming on and sharing. 791 00:42:12,920 --> 00:42:15,200 This was a blast. Thank you for having me. 792 00:42:15,560 --> 00:42:16,680 Cool. We'll round it off here. 793 00:42:16,680 --> 00:42:18,400 Then, if you're still listening, leave a like. 794 00:42:18,400 --> 00:42:20,440 If you like the episode, let us know in the comments section 795 00:42:20,440 --> 00:42:22,440 what you thought and we'll see you on the next one.