1 00:00:00,030 --> 00:00:01,860 AI tools are very bad at refactoring. 2 00:00:01,920 --> 00:00:03,750 They make mistakes. They introduce bugs. 3 00:00:03,810 --> 00:00:07,080 And they forget things or they introduce dead code. 4 00:00:07,110 --> 00:00:09,660 I advise against just producing huge amounts of code that you 5 00:00:09,660 --> 00:00:11,874 don't really understand. That doesn't seem like a recipe 6 00:00:11,874 --> 00:00:14,204 for success. Emily Bache, a technical coach 7 00:00:14,204 --> 00:00:16,654 and a member of Dave Farley's Modern Software Engineering 8 00:00:16,654 --> 00:00:19,446 channel, reveals how to level up your engineering teams in the 9 00:00:19,446 --> 00:00:22,129 age of AI. The more skilled you are with 10 00:00:22,129 --> 00:00:24,851 modern software engineering practices like working in small 11 00:00:24,851 --> 00:00:28,003 increments and writing tests. All of those skills are just 12 00:00:28,003 --> 00:00:29,725 amplified when you've got AI tools. 13 00:00:29,814 --> 00:00:32,034 How do you define the role of a technical coach? 14 00:00:32,064 --> 00:00:33,955 My job is not to deliver code myself. 15 00:00:33,985 --> 00:00:37,261 It's to deliver a team with a better capability to deliver 16 00:00:37,261 --> 00:00:40,006 better code. The two main activities I do are 17 00:00:40,006 --> 00:00:42,811 learning hours and ensemble. Everyone struggles with code 18 00:00:42,811 --> 00:00:44,741 quality. Everyone has a certain amount of 19 00:00:44,741 --> 00:00:46,510 technical debt. Everyone I've seen could learn 20 00:00:46,510 --> 00:00:48,205 to use the tools they have better. 21 00:00:48,265 --> 00:00:51,584 You don't learn how to produce beautiful code only by looking 22 00:00:51,584 --> 00:00:53,475 at a good design and analyzing it. 23 00:00:53,505 --> 00:00:57,255 Junior developers need mentoring to turn into senior developers. 24 00:00:57,855 --> 00:01:01,046 And mentoring is more than just writing comments and pull 25 00:01:01,046 --> 00:01:03,380 requests. Ensemble is another word for mob 26 00:01:03,380 --> 00:01:05,391 programming. You get a group of software 27 00:01:05,391 --> 00:01:07,425 developers together to collaborate and produce code. 28 00:01:07,455 --> 00:01:10,684 It's a great forum for learning because you see each other not 29 00:01:10,684 --> 00:01:12,774 only producing a code at the end. 30 00:01:12,774 --> 00:01:14,854 You get to see how people produce the code. 31 00:01:14,910 --> 00:01:17,010 There are two main failure modes for an ensemble. 32 00:01:17,010 --> 00:01:19,438 One is that you've got a group of five or six people and you've 33 00:01:19,438 --> 00:01:21,794 got one person with a keyboard and they're doing all the work, 34 00:01:21,794 --> 00:01:23,250 and everyone else is just zoning out. 35 00:01:23,430 --> 00:01:42,055 The other failure mode is. Hello, everyone. 36 00:01:42,055 --> 00:01:44,723 Welcome back to another new episode of the Tech Lead Journal 37 00:01:44,723 --> 00:01:46,474 podcast. Today, I have with me Emily 38 00:01:46,474 --> 00:01:48,625 Bache. Uh, she's actually a technical 39 00:01:48,625 --> 00:01:51,111 coach. Maybe some of you know technical 40 00:01:51,111 --> 00:01:53,218 coach, in your experience, in your career. 41 00:01:53,548 --> 00:01:56,601 But rarely actually I hear about technical coach, at least 42 00:01:56,601 --> 00:01:58,586 within, you know, my region, Southeast Asia. 43 00:01:59,121 --> 00:02:03,578 Especially Emily is specializing in this method called Samman 44 00:02:03,578 --> 00:02:05,791 method. Actually I have never heard 45 00:02:05,791 --> 00:02:08,122 about Samman method before until I did my research. 46 00:02:08,583 --> 00:02:11,242 So yeah, hopefully we can learn a thing or two about Samman 47 00:02:11,242 --> 00:02:14,654 method, why it is something that she's advocating, and we'll also 48 00:02:14,654 --> 00:02:17,690 learn about what a technical coach role is. 49 00:02:17,720 --> 00:02:20,828 So Emily, welcome to the show. Thank you for having me. 50 00:02:20,828 --> 00:02:23,183 I'm pleased to be invited. Right. 51 00:02:23,183 --> 00:02:25,853 Emily, I always love to invite my guests to probably share a 52 00:02:25,853 --> 00:02:27,503 little bit more from your career. 53 00:02:27,503 --> 00:02:29,933 Any turning points that you think we can learn from you? 54 00:02:31,156 --> 00:02:35,393 Yeah, so I've been in this business for like 25 years since 55 00:02:35,393 --> 00:02:38,169 I graduated with a degree in software engineering. 56 00:02:38,919 --> 00:02:43,368 And I, uh, we lived in the UK. And then I worked in the UK as a 57 00:02:43,368 --> 00:02:46,093 software developer for a bit. And then my husband and I moved 58 00:02:46,093 --> 00:02:47,949 to Sweden, and we've been here ever since. 59 00:02:48,658 --> 00:02:52,434 I was also very privileged to be an early adopter of extreme 60 00:02:52,434 --> 00:02:54,950 programming. I started doing that on a 61 00:02:54,950 --> 00:02:58,748 project in like the year 2000, and it was a huge influence on 62 00:02:58,748 --> 00:03:01,334 me. I've been doing like test-driven 63 00:03:01,334 --> 00:03:03,870 development and all that stuff for a long time now. 64 00:03:04,657 --> 00:03:08,344 Over the years, I've been trying to help others to learn these 65 00:03:08,344 --> 00:03:13,047 skills in various ways. I had a stint as an architect, 66 00:03:13,047 --> 00:03:16,597 test architect in a company. But these days I'm an 67 00:03:16,597 --> 00:03:19,573 independent consultant. And I technical coach and I go 68 00:03:19,573 --> 00:03:22,039 around different companies helping teams to learn better 69 00:03:22,039 --> 00:03:24,368 coding skills, basically. Yeah. 70 00:03:24,368 --> 00:03:27,338 And I saw you're actually also part of the Modern Engineering, 71 00:03:27,338 --> 00:03:30,732 so the, I dunno, the YouTube channel or the group set up by 72 00:03:30,732 --> 00:03:32,548 Dave Farley. So tell us a little bit more 73 00:03:32,548 --> 00:03:33,862 about that. Like what's your experience so 74 00:03:33,862 --> 00:03:35,602 far? Yeah. 75 00:03:35,602 --> 00:03:39,057 Oh, that's, that was, uh, I was so pleased when Dave Farley came 76 00:03:39,057 --> 00:03:42,535 to me and invited me to be one of the presenters on this, this 77 00:03:42,535 --> 00:03:44,507 channel. Um, it's really, it's really 78 00:03:44,507 --> 00:03:47,392 cool to get. Cause I, you know, I have ideas 79 00:03:47,392 --> 00:03:50,928 and stuff I wanna share and talk about, and Dave has built up 80 00:03:50,928 --> 00:03:55,202 this huge following on YouTube. So it's great to be able to talk 81 00:03:55,202 --> 00:03:57,122 about that stuff for that audience. 82 00:03:57,182 --> 00:04:00,466 But also that when I'm preparing what I'm gonna say, of course, 83 00:04:00,466 --> 00:04:03,440 Dave Farley gives me advice, you know, feedback on, on what I'm 84 00:04:03,440 --> 00:04:06,371 planning to say. And we have a chat amongst the 85 00:04:06,371 --> 00:04:08,570 presenters about what we wanna talk about. 86 00:04:08,570 --> 00:04:11,036 And it's really nice group to be involved with. 87 00:04:11,186 --> 00:04:14,696 So I hope people like to see the things that we produce as well. 88 00:04:15,731 --> 00:04:18,836 Yeah, personally, I've been seeing a couple of more, you 89 00:04:18,836 --> 00:04:22,400 know, contents either through email or videos, right, on 90 00:04:22,400 --> 00:04:25,035 YouTube. So hopefully we can see a lot of 91 00:04:25,035 --> 00:04:27,497 cool stuff happening from that Modern Engineering channel. 92 00:04:27,887 --> 00:04:31,111 So today we are going to talk about technical coaching with 93 00:04:31,111 --> 00:04:33,003 Samman methods. So maybe let's start with the 94 00:04:33,003 --> 00:04:34,717 role itself. Like I mentioned in the 95 00:04:34,717 --> 00:04:37,698 beginning, for some of us here within this region, it is 96 00:04:37,698 --> 00:04:40,946 actually very rare to hear about a team having technical coach. 97 00:04:41,276 --> 00:04:44,270 Maybe it's more prominent to have Agile coaches or other type 98 00:04:44,270 --> 00:04:46,556 of coaches like leadership coach and things like that. 99 00:04:46,836 --> 00:04:50,254 So maybe first thing, how do you define the role of a technical 100 00:04:50,254 --> 00:04:53,993 coach? So Agile coaches are more 101 00:04:53,993 --> 00:04:58,496 concerned with kind of teams and process and introducing Agile 102 00:04:58,496 --> 00:05:03,633 ceremonies and ways of working. But technical coach, my concern 103 00:05:03,633 --> 00:05:07,017 is much more around the code, the people writing the code, how 104 00:05:07,017 --> 00:05:10,397 the code is being written and whether we can raise the quality 105 00:05:10,397 --> 00:05:13,842 of the code and raise the skills of the people writing it. 106 00:05:14,381 --> 00:05:17,513 It's much more focused around that part of what's going on. 107 00:05:17,903 --> 00:05:20,063 And my background, as I said, is a software developer. 108 00:05:20,063 --> 00:05:24,117 So you need very good technical skills to be a technical coach. 109 00:05:24,207 --> 00:05:26,347 Uh, you need a background in software development. 110 00:05:26,847 --> 00:05:28,827 And not all Agile coaches have that. 111 00:05:29,247 --> 00:05:32,081 So it's, I see it as a complimentary role. 112 00:05:32,111 --> 00:05:35,151 I mean, I'd love to work with an Agile coach with a team. 113 00:05:35,631 --> 00:05:38,605 So I would get to focus more on the technical side of things and 114 00:05:38,605 --> 00:05:41,181 the Agile coach more on the process and the organization 115 00:05:41,181 --> 00:05:44,608 side of things. Yeah, probably if I can map it 116 00:05:44,608 --> 00:05:46,903 to like a common engineering team set up, right? 117 00:05:47,023 --> 00:05:49,553 So there's an engineering manager who is more on the 118 00:05:49,553 --> 00:05:52,431 process, you know, delivery and stuff while the technical lead 119 00:05:52,431 --> 00:05:54,608 probably is playing that role, you know. 120 00:05:54,608 --> 00:05:57,587 Like making sure that the code is written well, the process, 121 00:05:57,587 --> 00:05:59,871 engineering process. And you mentioned a couple of 122 00:05:59,871 --> 00:06:01,887 key things that I think are interesting, right? 123 00:06:01,887 --> 00:06:05,683 So you focus more on the code, uh, writing the code, you know, 124 00:06:05,683 --> 00:06:07,975 the skills to write a good code, I suppose. 125 00:06:08,245 --> 00:06:10,631 And the quality. So obviously these days there 126 00:06:10,631 --> 00:06:13,631 are a lot of changes in terms of trends and technology 127 00:06:13,631 --> 00:06:16,151 advancement and AI come into the picture, right? 128 00:06:16,525 --> 00:06:19,753 So what do you think? Is AI a disruptor to your role? 129 00:06:19,873 --> 00:06:23,347 Is AI an amplifier to your role? What has been your experience so 130 00:06:23,347 --> 00:06:26,969 far? Yeah, I mean the AI tools that 131 00:06:26,969 --> 00:06:31,009 are coming in are more and more impressive, and more and more 132 00:06:31,009 --> 00:06:35,206 people are are using them. So of course, my role is to help 133 00:06:35,206 --> 00:06:37,225 software developers to write better code. 134 00:06:37,255 --> 00:06:40,473 And the software developers that I work with all have these 135 00:06:40,473 --> 00:06:42,367 tools. They all have AI tools that 136 00:06:42,367 --> 00:06:45,259 they're starting to use. So I don't think it really 137 00:06:45,259 --> 00:06:48,419 changes my role so much as I also need to understand how to 138 00:06:48,419 --> 00:06:51,707 use these tools well so that I can help others with that. 139 00:06:52,187 --> 00:06:55,865 Because I think it's, I dunno if it's kind of this feeling that 140 00:06:55,865 --> 00:06:59,147 these AI tools are so easy to use that it doesn't need any 141 00:06:59,147 --> 00:07:01,745 training, that you just give them to all the developers and 142 00:07:01,745 --> 00:07:05,471 everyone will just work it out. But I think a lot of software 143 00:07:05,471 --> 00:07:07,923 developers actually benefit from getting some advice about how to 144 00:07:07,923 --> 00:07:11,372 best use these tools. And a lot of the advice that I'm 145 00:07:11,372 --> 00:07:14,936 giving at the moment is that, you know, it's that the more 146 00:07:14,936 --> 00:07:17,995 skilled you are with modern software engineering practices, 147 00:07:17,995 --> 00:07:22,587 like working in small increments and writing tests and being able 148 00:07:22,587 --> 00:07:27,707 to explain what the user benefit of the software is in tests. 149 00:07:27,887 --> 00:07:30,972 You know, all of those skills that, you know, I've been 150 00:07:30,972 --> 00:07:34,576 teaching for a long time are just amplified when you've got 151 00:07:34,576 --> 00:07:37,022 AI tools. And the AI tools work much 152 00:07:37,022 --> 00:07:39,022 better if you can do all those things. 153 00:07:39,592 --> 00:07:42,881 So I'm not sure that it's disrupted my role so much as I 154 00:07:42,881 --> 00:07:45,767 have to be able to incorporate advice about using these tools 155 00:07:45,767 --> 00:07:47,636 into the training and coaching I do. 156 00:07:48,500 --> 00:07:50,630 Yeah, personally, I fully agree with what you said, right? 157 00:07:50,630 --> 00:07:53,197 So given like a good understanding about the 158 00:07:53,197 --> 00:07:55,418 fundamentals, the practices, the important practices, like you 159 00:07:55,418 --> 00:07:58,232 mentioned, you know. The XP practices, for example, 160 00:07:58,232 --> 00:08:01,079 like test-driven development, having automated tests, the 161 00:08:01,079 --> 00:08:03,527 CI/CD integration, and things like that. 162 00:08:03,827 --> 00:08:07,482 Definitely using AI is going to amplify all those important 163 00:08:07,482 --> 00:08:11,347 stuff and you probably will get more things done in a higher 164 00:08:11,347 --> 00:08:13,912 quality. So that will be the wish of 165 00:08:13,912 --> 00:08:15,522 almost everybody in software engineering. 166 00:08:15,809 --> 00:08:18,112 But yeah, we'll see. I mean, whether AI can actually 167 00:08:18,112 --> 00:08:20,272 help people in terms of improving the quality. 168 00:08:20,872 --> 00:08:23,562 But one thing for sure about software engineering that I 169 00:08:23,562 --> 00:08:26,711 picked from reading the book, right, so it is mentioned that 170 00:08:26,711 --> 00:08:29,624 software engineering is actually more like a learning process 171 00:08:29,624 --> 00:08:32,958 rather than so-called, you know, doing the production process or, 172 00:08:32,958 --> 00:08:35,975 you know, just writing the code in order to build something, 173 00:08:35,975 --> 00:08:38,601 right? So tell us the important thing 174 00:08:38,601 --> 00:08:42,015 about this learning process. Yeah, I totally agree. 175 00:08:42,015 --> 00:08:44,996 I mean, I think this is what Dave Farley says in his book, 176 00:08:44,996 --> 00:08:47,290 Modern Software Engineering, that there's two big problems 177 00:08:47,290 --> 00:08:50,859 that you need to solve. And one is to manage the 178 00:08:50,859 --> 00:08:53,724 complexity in the software, in the world, in, you know, 179 00:08:53,724 --> 00:08:56,309 actually solving the problem, which this problem the software 180 00:08:56,309 --> 00:08:59,832 is solving is really complex. But the other half of it is 181 00:08:59,832 --> 00:09:02,480 optimizing for learning, because the world's changing and you 182 00:09:02,480 --> 00:09:05,592 need to be able to keep up. So these are the primary drivers 183 00:09:05,592 --> 00:09:08,925 of most of what we do in, when we're building software, we're 184 00:09:08,925 --> 00:09:12,762 trying to make sure that we are learning about what we need to 185 00:09:12,762 --> 00:09:15,743 build and we're trying to manage that complexity with good design 186 00:09:15,743 --> 00:09:19,636 and so on in the code. So yeah, I'm absolutely, uh, 187 00:09:19,636 --> 00:09:23,196 trying to help teams with that. Just to go back to, you said 188 00:09:23,196 --> 00:09:25,283 earlier about, you know, that it's similar, what I do is 189 00:09:25,283 --> 00:09:28,361 similar to a tech lead. So a tech lead would be on a 190 00:09:28,361 --> 00:09:30,546 team and they would be responsible for helping everyone 191 00:09:30,546 --> 00:09:32,603 around them to produce good code. 192 00:09:32,603 --> 00:09:34,578 And that is of course what I'm doing. 193 00:09:34,578 --> 00:09:37,144 But my job is not to deliver code myself. 194 00:09:37,534 --> 00:09:41,824 It's to deliver a team with a better capability to deliver 195 00:09:41,824 --> 00:09:43,753 better code. I'm a coach. 196 00:09:43,753 --> 00:09:46,949 I'm not delivering myself. I'm delivering, uh, improvements 197 00:09:46,949 --> 00:09:49,913 to other people. So it's a little bit meta in 198 00:09:49,913 --> 00:09:52,381 that sense. But of course I need to help 199 00:09:52,381 --> 00:09:54,649 them to optimize for learning and manage the complexity they 200 00:09:54,649 --> 00:09:56,227 have. Yeah. 201 00:09:56,317 --> 00:09:58,919 When you say optimize for learning for software engineers, 202 00:09:58,919 --> 00:10:01,417 what will be the typical things that you do? 203 00:10:01,611 --> 00:10:05,007 Or maybe this is also a good segue to what you are, you know, 204 00:10:05,007 --> 00:10:07,180 the Samman method that you are conducting as well. 205 00:10:08,623 --> 00:10:11,503 Yeah, so let's just outline the Samman method and how it can 206 00:10:11,503 --> 00:10:13,730 help with these aspects of software engineering. 207 00:10:13,730 --> 00:10:17,278 So, when I come in with a team, of course, I need to get to know 208 00:10:17,278 --> 00:10:20,337 that situation a little bit and talk to them and establish 209 00:10:20,337 --> 00:10:24,116 whether it's a good fit. But after that, the two main 210 00:10:24,116 --> 00:10:27,267 activities I do are learning hours and ensemble. 211 00:10:28,004 --> 00:10:31,440 The learning hours are where we kind of focus on building 212 00:10:31,440 --> 00:10:33,984 skills, doing exercises, practicing techniques that are 213 00:10:33,984 --> 00:10:36,670 gonna be useful in the production code. 214 00:10:37,420 --> 00:10:41,035 And then in the, uh, the main ensemble, that's where we go 215 00:10:41,035 --> 00:10:44,485 into their production code. I sit with the team and we try 216 00:10:44,485 --> 00:10:48,296 and, you know, build a feature or add some unit tests or do 217 00:10:48,296 --> 00:10:51,766 something realistic in their code and I'll coach them through 218 00:10:51,766 --> 00:10:55,143 it. And all the time I'm, so for 219 00:10:55,143 --> 00:10:58,028 things like optimize for learning, I mean, that's, you're 220 00:10:58,028 --> 00:11:00,652 trying to learn about the problem your software is trying 221 00:11:00,652 --> 00:11:04,736 to solve and... But also you need to be learning 222 00:11:04,736 --> 00:11:08,102 skills and improving and leveling up your ability to 223 00:11:08,102 --> 00:11:10,877 deliver. So I guess that's the aspect of 224 00:11:10,877 --> 00:11:13,007 optimizing for learning that I'm mostly focused on. 225 00:11:13,887 --> 00:11:16,023 And then the managing complexity, that's all about 226 00:11:16,023 --> 00:11:19,840 having design skills, basically, about being able to divide up 227 00:11:19,840 --> 00:11:23,170 the problem and decompose it into modules or classes. 228 00:11:23,560 --> 00:11:29,245 And also that the skills to evolve that design, to manage as 229 00:11:29,245 --> 00:11:31,982 the complexity increases, new features come in that you can 230 00:11:31,982 --> 00:11:34,756 evolve the design. And that's something that comes 231 00:11:34,756 --> 00:11:38,095 up in, uh, the skills that we learn in the learning hours, and 232 00:11:38,095 --> 00:11:40,119 of course the hands-on in the teams code. 233 00:11:40,942 --> 00:11:43,528 Yeah, learning hours and ensemble seems to be like the 234 00:11:43,528 --> 00:11:45,904 two core parts of the Samman method, right? 235 00:11:46,204 --> 00:11:50,098 So I guess, let's probably dive into the ensemble first, because 236 00:11:50,098 --> 00:11:52,933 I think this might be more interesting as well to discuss, 237 00:11:52,933 --> 00:11:55,198 right? So the way I read it, ensemble, 238 00:11:55,198 --> 00:11:57,510 right? Some people associate that with 239 00:11:57,510 --> 00:12:00,170 like mob programming, like programming together in a group. 240 00:12:00,410 --> 00:12:02,480 First of all, like how big should ensemble be? 241 00:12:02,480 --> 00:12:04,850 Because pair programming, is it considered ensemble? 242 00:12:05,150 --> 00:12:09,600 And like also, I'm sure many people haven't really tried mob 243 00:12:09,600 --> 00:12:13,331 programming or ensemble. And what will be your, you know, 244 00:12:13,331 --> 00:12:15,660 maybe elevator pitch for us to actually give it a try? 245 00:12:16,731 --> 00:12:19,666 Well, that was a lot of things. So, just to clarify, ensemble is 246 00:12:19,666 --> 00:12:21,187 another word for mob programming. 247 00:12:21,187 --> 00:12:24,812 It's exactly the same thing. And, uh, software teaming again, 248 00:12:24,812 --> 00:12:28,267 is just different words. Different people talk about the 249 00:12:28,267 --> 00:12:31,030 same activity. But the basic idea is that you 250 00:12:31,030 --> 00:12:33,119 get a group of software developers together to 251 00:12:33,119 --> 00:12:37,703 collaborate and produce code. And so it can be anything from 252 00:12:37,703 --> 00:12:41,598 three people upwards, basically. Normally, when I'm working with 253 00:12:41,598 --> 00:12:46,238 a team, it's a, it's the normal, normal Scrum team kind of size. 254 00:12:46,238 --> 00:12:51,218 So it's about 4, 5, 6 people. Those people are usually 255 00:12:51,218 --> 00:12:54,288 software developers. Quite often actually, we get 256 00:12:54,288 --> 00:12:56,630 joined by a tester. Not always. 257 00:12:56,870 --> 00:13:00,287 They don't always like to join us, but we are all the people in 258 00:13:00,287 --> 00:13:02,185 the team who are gonna write code together. 259 00:13:02,707 --> 00:13:05,448 Then you wanted also an elevator pitch, why would you do this, 260 00:13:05,448 --> 00:13:07,704 you know? 'Cause the, the normal way is of 261 00:13:07,704 --> 00:13:10,924 course, for each person to have their own computer and to work 262 00:13:10,924 --> 00:13:14,169 individually. But they, there are a number of 263 00:13:14,169 --> 00:13:15,804 advantages to collaborating like this. 264 00:13:16,194 --> 00:13:20,274 And one of them is that it's a great forum for learning, 265 00:13:20,274 --> 00:13:24,785 because you see each other not only producing a code at the 266 00:13:24,785 --> 00:13:26,748 end. You get to see how people 267 00:13:26,748 --> 00:13:30,245 produce the code. So just at, in the first 268 00:13:30,245 --> 00:13:33,385 instance, people can say, oh, your IDE does that. 269 00:13:33,385 --> 00:13:36,710 I didn't know you could do that. And people learn, just pick up 270 00:13:36,710 --> 00:13:38,155 like keyboard shortcuts and stuff. 271 00:13:38,691 --> 00:13:42,421 And then later, you know, just this whole thing of if you do a 272 00:13:42,421 --> 00:13:44,487 code review, you just see the end result. 273 00:13:44,844 --> 00:13:46,931 If you collaborate with someone, you see how they do it. 274 00:13:47,511 --> 00:13:51,329 And junior developers in particular will find that a good 275 00:13:51,329 --> 00:13:53,461 way for them to learn and level up. 276 00:13:54,293 --> 00:13:56,440 So it's a good forum for leveling up a team. 277 00:13:57,060 --> 00:14:01,121 It's also a good forum for, you know, as with pair programming, 278 00:14:01,271 --> 00:14:03,265 reducing the lottery ticket number. 279 00:14:03,715 --> 00:14:07,725 You know, if one person wins the lottery and suddenly disappears, 280 00:14:07,725 --> 00:14:10,941 how would you manage, you know, have they got so much knowledge 281 00:14:10,941 --> 00:14:13,418 that they've taken with them that the whole thing falls 282 00:14:13,418 --> 00:14:15,675 apart? So by doing this, we're sharing 283 00:14:15,675 --> 00:14:19,278 our knowledge and our context so that we are less reliant on 284 00:14:19,278 --> 00:14:22,167 individuals. So I think those are two of the 285 00:14:22,167 --> 00:14:24,029 main benefits of this style of working. 286 00:14:24,542 --> 00:14:28,357 But also, you know, once you get into it, I think it's quite a 287 00:14:28,357 --> 00:14:31,686 fun way to work actually. This kind of direct 288 00:14:31,686 --> 00:14:34,460 collaboration, where you're supporting one another in real 289 00:14:34,460 --> 00:14:36,019 time. Yeah. 290 00:14:36,049 --> 00:14:40,091 So unfortunately for me, my experience so far have been only 291 00:14:40,121 --> 00:14:41,861 maybe pair programming more, right? 292 00:14:42,101 --> 00:14:45,401 I haven't really developed in a, like a bigger group, right? 293 00:14:45,611 --> 00:14:48,275 And in fact if I, I understand correctly from Woody, right, the 294 00:14:48,275 --> 00:14:51,268 one who created the mob programming, he actually 295 00:14:51,268 --> 00:14:53,464 advocate technical people, non-technical people, including 296 00:14:53,464 --> 00:14:56,461 like business sometimes to just sit together and talk through 297 00:14:56,461 --> 00:14:58,606 the problems and actually come up with a solution, right? 298 00:14:59,056 --> 00:15:02,202 And also funnily enough in my career, I've experienced before 299 00:15:02,202 --> 00:15:05,485 like having to, you know, face a manager that doesn't like this 300 00:15:05,485 --> 00:15:08,288 kind of working. And I'm sure in many companies 301 00:15:08,288 --> 00:15:12,064 they also might feel that, you know, like a few people working 302 00:15:12,064 --> 00:15:15,373 together on, uh, on the same problem might be deemed as a 303 00:15:15,373 --> 00:15:18,376 waste of time, effort, or man hours, whatever that is you 304 00:15:18,376 --> 00:15:22,277 wanna call it, right? So how do you, first of all, you 305 00:15:22,277 --> 00:15:25,257 know, stand by your argument that this is beneficial? 306 00:15:25,347 --> 00:15:28,099 And how do you actually quantify that for those kind of managers 307 00:15:28,099 --> 00:15:30,027 who still don't see it as a valuable thing? 308 00:15:30,942 --> 00:15:33,884 Yeah, I don't know. I'm interested in kind of 309 00:15:33,884 --> 00:15:37,115 browbeating any managers. But I think you, at some level, 310 00:15:37,115 --> 00:15:39,395 need to trust the team that they know what they're doing. 311 00:15:40,385 --> 00:15:43,256 And if they wanna work that way, then let them. 312 00:15:43,536 --> 00:15:46,490 And see how it goes. But I just to come back to this 313 00:15:46,490 --> 00:15:49,556 thing about, you know, the normal way is to divide up the 314 00:15:49,556 --> 00:15:54,169 work and to have handovers. And every time you have a 315 00:15:54,169 --> 00:15:58,360 handover of a piece of work from one role to another, there's a 316 00:15:58,360 --> 00:16:02,149 potential for something to get lost and for time to get wasted 317 00:16:02,149 --> 00:16:05,316 and context to be lost. And then, so the thing with 318 00:16:05,316 --> 00:16:08,204 getting everyone in the same room in, like as Woody says 319 00:16:08,204 --> 00:16:11,467 including business people and testers, you get this 320 00:16:11,467 --> 00:16:14,720 possibility to have single piece flow where you take one task, 321 00:16:14,720 --> 00:16:18,188 you've got all the people in the room who are capable of taking 322 00:16:18,188 --> 00:16:21,058 this task from start to finish and out into production. 323 00:16:21,798 --> 00:16:25,316 And without any kind of handovers in a way because 324 00:16:25,316 --> 00:16:27,398 everyone is present for all of the tasks. 325 00:16:27,680 --> 00:16:30,082 You eliminate these opportunities for 326 00:16:30,082 --> 00:16:32,212 misunderstandings and context to get dropped. 327 00:16:32,697 --> 00:16:35,995 The result that I observe is that the overall quality of the 328 00:16:35,995 --> 00:16:39,375 thing you produce is higher. And actually you get it sooner 329 00:16:39,375 --> 00:16:42,595 because you didn't have to wait for any of those handovers. 330 00:16:43,145 --> 00:16:47,018 So I suggest to the manager that instead of trying to look at 331 00:16:47,018 --> 00:16:52,165 person utilization, they start to look at flow and the flow of 332 00:16:52,165 --> 00:16:54,122 features. It's a very lean concept, you 333 00:16:54,122 --> 00:16:56,057 know. I think that's the way to look 334 00:16:56,057 --> 00:16:58,097 at it. It's a way to get single piece 335 00:16:58,097 --> 00:17:01,238 flow in your development. I like that you mentioned about 336 00:17:01,238 --> 00:17:02,616 flow, right? Single piece flow. 337 00:17:02,616 --> 00:17:05,195 I think that's very crucial in the lean methodology, right? 338 00:17:05,586 --> 00:17:08,096 And yeah, these days in many software engineering team, I 339 00:17:08,096 --> 00:17:10,373 would say that there are specialization, there are roles, 340 00:17:10,373 --> 00:17:12,786 right? Maybe these days is getting much 341 00:17:12,786 --> 00:17:15,701 more blended because of the AI introduction, right? 342 00:17:15,701 --> 00:17:18,979 So like a developer now can do maybe maybe other different 343 00:17:18,979 --> 00:17:21,691 roles altogether as well. But yeah, that single piece of 344 00:17:21,691 --> 00:17:24,256 flow is something that probably as a manager, right, you should 345 00:17:24,256 --> 00:17:25,786 try to optimize much more, right? 346 00:17:26,266 --> 00:17:28,666 So maybe if we go to the mechanics first, right? 347 00:17:28,666 --> 00:17:31,636 Because maybe people are interested in giving a try. 348 00:17:31,798 --> 00:17:34,103 I mean it should not be just, you know, a group of people just 349 00:17:34,103 --> 00:17:37,463 gather on one, I dunno, monitor or one laptop and just talk. 350 00:17:38,029 --> 00:17:40,967 What are some of the mechanics that you think people can have a 351 00:17:40,967 --> 00:17:42,595 guide? Maybe it doesn't have to be so 352 00:17:42,595 --> 00:17:45,230 deep, but at least they have a good sense how to give it a try. 353 00:17:46,051 --> 00:17:49,094 Yeah, yeah. So if you wanna give it a try, 354 00:17:49,094 --> 00:17:52,216 then I would advise trying to get help from somebody who's 355 00:17:52,216 --> 00:17:55,194 done it before, because it's quite easy to fall into one of 356 00:17:55,194 --> 00:17:58,554 the failure modes with this. There are two main failure modes 357 00:17:58,554 --> 00:18:00,727 for an ensemble. One is that everyone, you've got 358 00:18:00,727 --> 00:18:03,709 a group of what, five or six people and you've got one person 359 00:18:03,709 --> 00:18:06,340 with a keyboard, and they're doing all the work and everyone 360 00:18:06,340 --> 00:18:09,406 else is just zoning out. And they're, you know, just 361 00:18:09,406 --> 00:18:11,611 checking their phone or looking out the window or whatever. 362 00:18:12,061 --> 00:18:14,841 So that's the first failure mode that, you know, there's only one 363 00:18:14,841 --> 00:18:17,491 person doing any work. That's not what we're after. 364 00:18:17,791 --> 00:18:20,856 The other failure mode is kind of similar but different. 365 00:18:20,856 --> 00:18:24,232 It's whether the person with a keyboard is sitting there and 366 00:18:24,232 --> 00:18:26,920 everyone else is shouting different things at them to do. 367 00:18:27,404 --> 00:18:29,574 They don't get anything done either because they basically 368 00:18:29,574 --> 00:18:32,032 have to ignore everyone and do what they wanted because 369 00:18:32,032 --> 00:18:33,267 everyone is giving conflicting advice. 370 00:18:33,646 --> 00:18:35,356 Those are the basic two failure modes. 371 00:18:35,596 --> 00:18:38,624 What you want is a collaboration where everyone is contributing 372 00:18:38,624 --> 00:18:40,486 to the software that's getting produced. 373 00:18:41,216 --> 00:18:46,559 And the how you achieve that is it takes some practice. 374 00:18:46,559 --> 00:18:48,029 It's a skill that you need to learn. 375 00:18:48,549 --> 00:18:53,067 And the way I normally teach it is we start with basically 376 00:18:53,067 --> 00:18:56,052 nominating somebody for the talking role and making sure 377 00:18:56,052 --> 00:18:59,397 that that role rotates through all the people who are present 378 00:18:59,397 --> 00:19:03,192 on a timer so that everyone gets a turn to say their thing. 379 00:19:03,432 --> 00:19:05,466 And you don't need to all shout at once 'cause you're gonna get 380 00:19:05,466 --> 00:19:07,848 your turn in a minute. And you don't zone out 'cause 381 00:19:07,848 --> 00:19:10,352 you know it's gonna be your turn in a minute, you know? 382 00:19:10,352 --> 00:19:12,714 So it's, that's the kind of one of the mechanisms. 383 00:19:12,714 --> 00:19:17,130 So separate the talking and the typing roles and rotate at least 384 00:19:17,130 --> 00:19:21,474 the talking roll on a scale of maybe five minutes per turn. 385 00:19:21,504 --> 00:19:24,382 Something like that. So sometimes I see also, 386 00:19:24,382 --> 00:19:26,985 especially in pair, right, they can come up with multiple 387 00:19:26,985 --> 00:19:31,137 keyboards together, you know, uh, I dunno, maybe you can even 388 00:19:31,137 --> 00:19:33,281 type different things on different keyboards, right? 389 00:19:33,281 --> 00:19:35,904 Is this something that should be avoided or is this something 390 00:19:35,904 --> 00:19:38,210 that can also help in terms of the ensemble? 391 00:19:39,242 --> 00:19:42,740 I think the normal, the normal flow of work is that everyone's 392 00:19:42,740 --> 00:19:44,621 attention is directed to the same screen. 393 00:19:45,101 --> 00:19:47,441 And having additional keyboards doesn't really help. 394 00:19:47,441 --> 00:19:51,781 'Cause you want everyone to be in the same task of paying 395 00:19:51,781 --> 00:19:54,808 attention to the same things. And the danger if you have 396 00:19:54,808 --> 00:19:57,440 several people typing like in different files, is that you 397 00:19:57,440 --> 00:20:00,992 basically just you are doing two tasks in the same code base 398 00:20:00,992 --> 00:20:03,422 instead of cooperating on one task. 399 00:20:03,940 --> 00:20:06,096 But of course everyone, when we're doing this online, 400 00:20:06,096 --> 00:20:08,169 everyone's got their own computer and it doesn't take 401 00:20:08,169 --> 00:20:11,334 long to just go and Google a thing and come back to the group 402 00:20:11,334 --> 00:20:14,415 and say, oh, I found this thing that's useful or I found the 403 00:20:14,415 --> 00:20:15,992 name of that method you're trying to call. 404 00:20:15,992 --> 00:20:19,299 Or you know, so you, people can do bits of research and bring 405 00:20:19,299 --> 00:20:23,101 that back to the group. But the, when you're working in 406 00:20:23,101 --> 00:20:26,499 ensemble, generally everyone is focused on the same activity. 407 00:20:27,525 --> 00:20:28,888 Yeah. And for that particular 408 00:20:28,888 --> 00:20:31,702 activity, maybe if you can give guidance typically, uh, is it 409 00:20:31,702 --> 00:20:35,468 something like you pick up a, I dunno, a ticket, uh, and then 410 00:20:35,468 --> 00:20:38,577 you finish it end to end. Like from the development, 411 00:20:38,577 --> 00:20:40,920 design, development, testing, deployment and all that? 412 00:20:41,100 --> 00:20:44,046 Or is there like a typical, I dunno, activity that you 413 00:20:44,046 --> 00:20:46,456 normally do? Is it like code kata or just 414 00:20:46,456 --> 00:20:48,060 introducing tests or something like that? 415 00:20:48,962 --> 00:20:51,867 Well, I think the ideal is that if you can get a task all the 416 00:20:51,867 --> 00:20:55,541 way from start to finish. But of course, you might not 417 00:20:55,541 --> 00:20:59,127 decide to do every aspect of it. I think you learn that from 418 00:20:59,127 --> 00:21:01,059 experience. What kinds of tasks work well 419 00:21:01,059 --> 00:21:04,026 and for what kinds of tasks the people you've got gathered 420 00:21:04,026 --> 00:21:06,667 aren't the right people. And you need to do something 421 00:21:06,667 --> 00:21:08,965 else. When you're first learning, I 422 00:21:08,965 --> 00:21:12,223 think code katas are useful to practice on, cause of obviously 423 00:21:12,223 --> 00:21:16,028 the, the amount of code that you need to understand is much 424 00:21:16,028 --> 00:21:19,659 smaller and you can focus more on how the interactions are 425 00:21:19,659 --> 00:21:21,329 happening and how you're communicating. 426 00:21:21,719 --> 00:21:24,712 So it's a good thing to do if you're just starting out to 427 00:21:24,712 --> 00:21:28,648 start with some code katas. I think if you read up on how 428 00:21:28,648 --> 00:21:31,441 Woody Zuill originally got his team doing this, it was, they 429 00:21:31,441 --> 00:21:33,868 did a lot of code katas before they tackled production code. 430 00:21:34,558 --> 00:21:38,375 So I think just getting the, uh, the communication patterns down 431 00:21:38,375 --> 00:21:41,999 when, in a situation where the stakes are low with the code, 432 00:21:41,999 --> 00:21:45,542 that can be a good thing. But it is a technique that is, 433 00:21:45,542 --> 00:21:48,654 you'll only get the real benefit of it if you're doing it in 434 00:21:48,654 --> 00:21:51,415 production code, I guess. Yeah, you have emphasized a 435 00:21:51,415 --> 00:21:53,395 couple of times about working on the production code. 436 00:21:53,425 --> 00:21:55,145 I guess working on code kata can be fun. 437 00:21:55,165 --> 00:21:56,515 And you can learn something, right? 438 00:21:56,515 --> 00:21:59,080 But there's nothing that can beat, you know, understanding 439 00:21:59,080 --> 00:22:01,075 production code, especially the complex one. 440 00:22:01,495 --> 00:22:03,736 You know, where everybody has probably different parts of 441 00:22:03,736 --> 00:22:05,215 knowledge within the code base, right? 442 00:22:05,455 --> 00:22:08,446 And you can get to learn from each other, the techniques, the 443 00:22:08,446 --> 00:22:11,063 design, and the issues that typically happen for some parts 444 00:22:11,063 --> 00:22:14,195 of the code base. So another important aspect in 445 00:22:14,195 --> 00:22:18,162 ensemble that I also read from the book is the importance of 446 00:22:18,162 --> 00:22:21,480 like, trust, communication skills, and also kindness, 447 00:22:21,480 --> 00:22:23,704 right? Because I'm sure everybody would 448 00:22:23,704 --> 00:22:27,580 have seen bad code in, you know, anybody's code base, right? 449 00:22:27,580 --> 00:22:30,370 So, and especially when working together, you can see different 450 00:22:30,370 --> 00:22:33,857 people's code, being written as the result of, I don't know, 451 00:22:33,857 --> 00:22:36,079 maybe sometime ago or maybe just recent, right? 452 00:22:36,409 --> 00:22:40,039 So any important points that you wanna convey about this aspect, 453 00:22:40,039 --> 00:22:42,529 like trust, communication, and kindness part. 454 00:22:43,689 --> 00:22:46,907 It's all about teamwork and it's about the communication that is 455 00:22:46,907 --> 00:22:50,638 gonna be respectful and helpful and, you know, get this 456 00:22:50,638 --> 00:22:53,010 collaboration going. And the person who wrote the 457 00:22:53,010 --> 00:22:54,882 code might be in the room, or they might not. 458 00:22:55,302 --> 00:22:58,128 But in either case, you need to talk about the code, as you say, 459 00:22:58,128 --> 00:23:00,815 in a respectful way. You don't know what they were 460 00:23:00,815 --> 00:23:03,701 suffering when they wrote it. It might look completely wrong 461 00:23:03,701 --> 00:23:05,969 and off the wall and badly written to you. 462 00:23:05,969 --> 00:23:08,423 But there's probably a reason for it to be like that. 463 00:23:08,423 --> 00:23:11,890 And you do need to be fine and respectful in the way you 464 00:23:11,890 --> 00:23:14,894 interact with one another and with, and talk about the code. 465 00:23:15,344 --> 00:23:17,980 So that's part of it as well, it's all about teamwork. 466 00:23:18,100 --> 00:23:19,570 And this is what you need in a team anyway. 467 00:23:19,570 --> 00:23:21,864 You've gotta trust your colleagues, you've gotta respect 468 00:23:21,864 --> 00:23:23,860 them. And that's what it's all about. 469 00:23:24,010 --> 00:23:27,917 A good ensemble just looks like a really, really gelled and good 470 00:23:27,917 --> 00:23:30,872 team. Yeah, and I like the quote that 471 00:23:30,872 --> 00:23:33,018 Woody mentioned about this, uh, ensemble thing, right? 472 00:23:33,018 --> 00:23:35,038 Like "honor the code and the coder", right? 473 00:23:35,038 --> 00:23:38,008 So of course you don't know what constraints they were dealing 474 00:23:38,008 --> 00:23:41,436 with when they wrote the code. This is also probably like in 475 00:23:41,436 --> 00:23:43,868 the agile retrospective, normally we call this prime 476 00:23:43,979 --> 00:23:45,231 directive or something like that, right? 477 00:23:45,231 --> 00:23:46,821 So no matter what happened, right? 478 00:23:47,181 --> 00:23:49,711 Maybe that was the best thing that happened at that particular 479 00:23:49,711 --> 00:23:51,532 point in time with the right person. 480 00:23:52,042 --> 00:23:54,992 So maybe from your experience, you have been technical coach 481 00:23:54,992 --> 00:23:57,742 for quite some time now and doing this ensemble. 482 00:23:57,742 --> 00:24:01,423 What are some of the cool things that you have, I dunno, observed 483 00:24:01,423 --> 00:24:04,222 or maybe you have achieved by doing this ensemble? 484 00:24:05,064 --> 00:24:06,758 Right. So, well, I have this privilege, 485 00:24:06,758 --> 00:24:09,852 I guess, that I get to go into a lot of different organizations 486 00:24:09,852 --> 00:24:13,869 and teams to coach. There are coaches who have a 487 00:24:13,869 --> 00:24:16,701 much more long-term relationship with like one organization and 488 00:24:16,701 --> 00:24:18,469 several teams in that organization. 489 00:24:19,399 --> 00:24:22,221 And I think that if the benefit of having this long-term 490 00:24:22,221 --> 00:24:25,213 relationship is you really get to see teams develop and improve 491 00:24:25,213 --> 00:24:27,779 over time. And I've seen a little bit of 492 00:24:27,779 --> 00:24:30,471 that happening with other coaches that I've worked with. 493 00:24:30,861 --> 00:24:35,351 But for me, I, I'm a bit of a, I hop from different organization 494 00:24:35,351 --> 00:24:38,020 to different organization and just get a short amount of 495 00:24:38,020 --> 00:24:41,131 coaching with lots of different teams and different places. 496 00:24:41,682 --> 00:24:44,230 I'm not quite sure how, why it's like that, but that seems to be 497 00:24:44,230 --> 00:24:45,942 the style of engagement that I have. 498 00:24:46,422 --> 00:24:50,203 And it means I get to see a huge variety of teams and 499 00:24:50,203 --> 00:24:53,373 organizations. And I guess I get to see that 500 00:24:53,373 --> 00:24:55,173 everyone struggles with code quality. 501 00:24:55,923 --> 00:24:58,173 Everyone has a certain amount of technical debt. 502 00:24:58,863 --> 00:25:02,085 And everyone, basically, everyone I've seen could learn 503 00:25:02,085 --> 00:25:04,209 to use the tools they have better. 504 00:25:04,989 --> 00:25:08,099 I mean, everyone's got these amazing refactoring tools and AI 505 00:25:08,099 --> 00:25:10,763 tools and all of that. And most developers are not 506 00:25:10,763 --> 00:25:12,869 using even a fraction of what it's capable of. 507 00:25:13,549 --> 00:25:16,841 So I get to have that reinforced every time I visit the team. 508 00:25:17,171 --> 00:25:18,809 Let me show you. There's all these things that 509 00:25:18,809 --> 00:25:21,985 you could be doing. I like to, uh, get people using 510 00:25:21,985 --> 00:25:23,671 more shortcuts and more, you know. 511 00:25:24,421 --> 00:25:27,493 I mean, most people have got rename these days, but no, 512 00:25:27,493 --> 00:25:30,311 introduce variable, introduce parameter, move method. 513 00:25:30,361 --> 00:25:32,681 These are refactorings that a lot of people have never tried. 514 00:25:33,331 --> 00:25:36,659 And so I'm doing a lot of just getting people to use their 515 00:25:36,659 --> 00:25:38,900 tools better. How about prompting? 516 00:25:39,050 --> 00:25:41,882 Prompt engineering? So have you, um, introduced more 517 00:25:41,882 --> 00:25:45,223 of that? Well, this is obviously quite a 518 00:25:45,223 --> 00:25:48,033 new thing for most teams to have these AI tools. 519 00:25:48,303 --> 00:25:52,255 So I have to say I haven't got a, a lot of hours under my belt 520 00:25:52,255 --> 00:25:55,313 yet working with them. It's, of course, the skills, as 521 00:25:55,313 --> 00:25:58,573 I said, are similar to the skills I know about, you know, 522 00:25:58,573 --> 00:26:02,420 working in small chunks. Giving the prompt for quite a 523 00:26:02,420 --> 00:26:05,424 well-defined task and making sure you know how to verify that 524 00:26:05,424 --> 00:26:06,796 the thing has done the right thing. 525 00:26:06,796 --> 00:26:11,163 So having tests and knowing how to write those tests and 526 00:26:11,163 --> 00:26:13,456 checking the diff for what it's actually done. 527 00:26:14,013 --> 00:26:17,836 All of these skills are things that maybe not completely 528 00:26:17,836 --> 00:26:20,750 obvious to everyone. And more and more with the teams 529 00:26:20,750 --> 00:26:22,402 I'm working with, we have to bring those in. 530 00:26:22,402 --> 00:26:26,009 And we start in the ensemble to the person talking is 531 00:26:26,009 --> 00:26:27,862 explaining. I think we need to prompt it 532 00:26:27,862 --> 00:26:31,025 like this and this is what I'm gonna intend to verify when it's 533 00:26:31,025 --> 00:26:33,811 done its thing. So there's a lot more of that 534 00:26:33,811 --> 00:26:36,650 coming in, but it's early days for me at the moment with all 535 00:26:36,650 --> 00:26:40,294 those AI tools, actually. As it is for many people, I 536 00:26:40,294 --> 00:26:42,401 think. Yeah, I'm sure many people will 537 00:26:42,401 --> 00:26:44,543 also feel it's early days for them, right? 538 00:26:44,543 --> 00:26:47,185 Picking up new tools. Not to mention there are new 539 00:26:47,185 --> 00:26:50,871 models coming in, new tools or new IDEs coming in, new agents 540 00:26:50,871 --> 00:26:53,386 being invented. So obviously, there's always new 541 00:26:53,386 --> 00:26:56,885 things to learn these days. And I like that the points that 542 00:26:56,885 --> 00:26:58,845 you mentioned about your experience, right, observing 543 00:26:58,845 --> 00:27:01,383 multiple teams, right? They're always code quality 544 00:27:01,383 --> 00:27:02,973 issue. They are always tech debt. 545 00:27:03,410 --> 00:27:05,323 And they're always underutilized tool. 546 00:27:05,563 --> 00:27:08,712 So I think, many software engineers, I believe could 547 00:27:08,712 --> 00:27:10,655 relate. So no matter where you go in any 548 00:27:10,655 --> 00:27:12,985 company, right, you will see these themes over and over 549 00:27:12,985 --> 00:27:14,400 again. So thanks for highlighting that. 550 00:27:14,520 --> 00:27:16,020 I think that is quite insightful. 551 00:27:16,860 --> 00:27:19,548 So the other aspect of Samman method that you mentioned is 552 00:27:19,548 --> 00:27:22,735 about learning hours. I think learning is kind of like 553 00:27:22,735 --> 00:27:24,538 given. Everybody knows that we have to 554 00:27:24,538 --> 00:27:26,552 learn. But what specific things about 555 00:27:26,552 --> 00:27:28,322 learning hours that is different? 556 00:27:29,443 --> 00:27:34,095 Well, I've been a, I was a very early adopter of the Coding Dojo 557 00:27:34,095 --> 00:27:37,319 back in the day, which is a forum where a bunch of 558 00:27:37,319 --> 00:27:40,427 developers get together to do code katas and increase skills. 559 00:27:41,027 --> 00:27:44,177 And I found that that was a really helpful thing to do. 560 00:27:44,600 --> 00:27:47,572 But over the years, I realized that you could make it even 561 00:27:47,572 --> 00:27:50,655 better by using some techniques from instructional design. 562 00:27:50,925 --> 00:27:54,885 And in particular, I was very inspired by Sharon Bowman, who 563 00:27:54,885 --> 00:27:57,664 is, uh, written this book Training from the Back of the 564 00:27:57,664 --> 00:27:59,889 Room. And she's nothing to do with 565 00:27:59,889 --> 00:28:02,699 software, but she knows a lot about instructional design and 566 00:28:02,699 --> 00:28:07,229 how to make learning experience is more sticky so that you 567 00:28:07,229 --> 00:28:10,149 remember more of the stuff and it has more of an impact in what 568 00:28:10,149 --> 00:28:12,902 you do afterwards. I've been very inspired to take 569 00:28:12,902 --> 00:28:16,424 a coding dojo, make it shorter, just an hour, you know, maybe a 570 00:28:16,424 --> 00:28:19,605 little bit longer than an hour, but still pretty short for a 571 00:28:19,605 --> 00:28:22,492 training. And trying to make sure that 572 00:28:22,492 --> 00:28:26,000 it's structured to take full advantage of how, what's gonna 573 00:28:26,000 --> 00:28:29,661 keep your attention? I use the 4C model, which comes 574 00:28:29,661 --> 00:28:33,611 from the book. Uh, we open with a connect, get 575 00:28:33,611 --> 00:28:35,731 everyone saying something about what they already know. 576 00:28:36,132 --> 00:28:39,642 We do a new concept, why I present what it is we're gonna 577 00:28:39,642 --> 00:28:41,801 learn, the new skill or the technique. 578 00:28:42,161 --> 00:28:44,641 Then we have some concrete practice where everyone does 579 00:28:44,641 --> 00:28:47,311 some coding. Usually coding exercise or some 580 00:28:47,311 --> 00:28:49,419 form of exercise. And then the conclusions at the 581 00:28:49,419 --> 00:28:52,363 end where we just try and make sure that everyone is, summarize 582 00:28:52,363 --> 00:28:56,124 in their own words what we've learned so that they can 583 00:28:56,124 --> 00:28:59,220 remember it better. So that's the structure that the 584 00:28:59,220 --> 00:29:03,194 learning hour has that I, so I designed these as a, you know, 585 00:29:03,194 --> 00:29:06,833 learning experiences for a team of software developers or to 586 00:29:06,833 --> 00:29:10,719 learn coding skills. So thanks for sharing about this 587 00:29:10,719 --> 00:29:12,532 Coding Dojo. I know that you have written a 588 00:29:12,532 --> 00:29:15,872 book about it as well, so I think in the past I, I learned a 589 00:29:15,872 --> 00:29:18,248 lot about this coding kata, Coding Dojo, right? 590 00:29:18,248 --> 00:29:21,110 People are doing, I think I haven't heard about it lately. 591 00:29:21,674 --> 00:29:23,555 I dunno why. Maybe people are dealing more 592 00:29:23,555 --> 00:29:25,443 with the AI stuff by themselves solo. 593 00:29:25,777 --> 00:29:28,160 But I think having this experience, learning in a dojo, 594 00:29:28,160 --> 00:29:31,034 like with the targeted focus of problem, right? 595 00:29:31,034 --> 00:29:33,673 So I think, with the right coach, which is also important, 596 00:29:33,673 --> 00:29:35,819 right, can give a lot of benefits, right? 597 00:29:36,089 --> 00:29:38,609 And you mentioned about specifically reducing it to an 598 00:29:38,609 --> 00:29:41,838 hour, is there any particular thing about that hour? 599 00:29:41,838 --> 00:29:45,411 Should we find an hour within our working days to actually do 600 00:29:45,411 --> 00:29:47,418 this? Or should we do it outside 601 00:29:47,418 --> 00:29:51,730 working hours? The Coding Dojo concept was a 602 00:29:51,730 --> 00:29:55,190 little bit based on the idea that you would spend your 603 00:29:55,190 --> 00:29:58,012 evenings, possibly your weekends, doing this kind of 604 00:29:58,012 --> 00:30:00,814 training. And I did, I put in evenings of 605 00:30:00,814 --> 00:30:04,602 weekends back in the day. And I found that, for me, that 606 00:30:04,602 --> 00:30:06,333 was possible. And I thrived on that. 607 00:30:06,333 --> 00:30:10,113 But for most developers, asking them to code in evenings and 608 00:30:10,113 --> 00:30:11,523 weekends, it's just not gonna happen. 609 00:30:11,523 --> 00:30:12,573 And they're just not gonna do it. 610 00:30:13,293 --> 00:30:16,353 And for the organization, that's a bit of a failure because you 611 00:30:16,353 --> 00:30:18,678 want people to upskill, learn stuff. 612 00:30:19,288 --> 00:30:22,522 And I think the best way to ensure that they put the time in 613 00:30:22,522 --> 00:30:24,613 is to give them time in the working week. 614 00:30:25,373 --> 00:30:28,604 And that's where another reason for making it shorter, because 615 00:30:28,604 --> 00:30:32,724 it's much less difficult to get a manager to agree to a one hour 616 00:30:32,724 --> 00:30:35,688 training session than to get them to agree to a half a day, 617 00:30:35,688 --> 00:30:38,074 you know? And you might be able to do it a 618 00:30:38,074 --> 00:30:40,408 bit more often and with practical skills like coding, 619 00:30:40,408 --> 00:30:44,561 the little and often is actually more effective than doing it 620 00:30:44,561 --> 00:30:48,037 occasionally for longer. Obviously, when I come in and do 621 00:30:48,037 --> 00:30:50,831 the coaching, the team will be doing learning hours with me for 622 00:30:50,831 --> 00:30:54,497 all the days that I'm there. But even when I'm not there, my 623 00:30:54,497 --> 00:30:58,211 hope is that they will perhaps be able to find an hour each, 624 00:30:58,211 --> 00:31:02,423 each iteration or sprint maybe to do one of these and or even 625 00:31:02,423 --> 00:31:04,337 more frequently if the boss will agree to it. 626 00:31:04,512 --> 00:31:07,992 But that it should be part of the normal work. 627 00:31:08,482 --> 00:31:11,350 Learning is part of your work, so let's put it into the 628 00:31:11,350 --> 00:31:14,435 schedule and make sure that it happens on a regular basis. 629 00:31:14,795 --> 00:31:16,283 That's the philosophy I think here. 630 00:31:17,298 --> 00:31:18,978 Yeah, so learning is part of the work. 631 00:31:18,978 --> 00:31:21,286 So I think this is very important, especially for 632 00:31:21,286 --> 00:31:23,018 knowledge worker. I think software engineers are 633 00:31:23,018 --> 00:31:24,738 knowledge worker, right? So we have to learn. 634 00:31:24,738 --> 00:31:26,490 In the beginning, we also mentioned that software 635 00:31:26,490 --> 00:31:28,188 engineering is a learning process, right? 636 00:31:28,489 --> 00:31:31,525 Tackling complexity. Making sure that we can learn 637 00:31:31,525 --> 00:31:34,254 the problem better. And I think you mentioned as 638 00:31:34,254 --> 00:31:37,193 well about learning, right? That we have to spend the time 639 00:31:37,193 --> 00:31:39,711 to actually learn. I don't know, like how people, 640 00:31:39,711 --> 00:31:42,221 uh, learn these days, right? I'm sure there are many 641 00:31:42,221 --> 00:31:44,406 channels, right? But learning as a team is 642 00:31:44,406 --> 00:31:47,318 probably a more powerful thing to do as well like the ensemble 643 00:31:47,318 --> 00:31:51,271 thing that you mentioned, right? So maybe these days people have 644 00:31:51,271 --> 00:31:53,413 instilled in their mind agile practices. 645 00:31:53,413 --> 00:31:56,653 Things like, I dunno, daily standup retrospective, demo, and 646 00:31:56,653 --> 00:32:00,191 things like that. What will be some typical, I 647 00:32:00,191 --> 00:32:03,269 dunno, ceremonies for learning that you advocate for your 648 00:32:03,269 --> 00:32:06,189 teams? Yeah, so I guess I was trying to 649 00:32:06,189 --> 00:32:09,270 hint there that if you can get a learning hour in each sprint, 650 00:32:09,270 --> 00:32:12,828 then that's a cadence that I think you should be able to 651 00:32:12,828 --> 00:32:15,219 sustain. If you can get more than that, 652 00:32:15,219 --> 00:32:18,804 then that's a bonus. What I see when I go into most 653 00:32:18,804 --> 00:32:21,840 teams is that the only learning activity they have is code 654 00:32:21,840 --> 00:32:24,480 reviews as part of the pull request process. 655 00:32:25,170 --> 00:32:28,807 And that is just universal. It seems to me everyone is doing 656 00:32:28,807 --> 00:32:31,113 these pull requests. And I'm, it's fine. 657 00:32:31,173 --> 00:32:34,264 Code review is a good thing to do, but it's happening after the 658 00:32:34,264 --> 00:32:36,775 code is written. It's happening asynchronously. 659 00:32:36,775 --> 00:32:40,757 It's written. And these are not great for 660 00:32:40,757 --> 00:32:42,722 actually training up junior developers. 661 00:32:43,082 --> 00:32:46,708 You want something a lot more kind of face-to-face interactive 662 00:32:46,708 --> 00:32:50,418 happening while the work is going on and to actually help 663 00:32:50,418 --> 00:32:53,961 them to upskill and to turn into those senior developers that you 664 00:32:53,961 --> 00:32:56,516 want. I think something like pair 665 00:32:56,516 --> 00:32:59,633 programming or ensemble is, that's the forum where you 666 00:32:59,633 --> 00:33:01,566 actually turn juniors into seniors. 667 00:33:02,249 --> 00:33:05,243 And you can do that obviously on the production code, so that you 668 00:33:05,243 --> 00:33:07,866 get work done at the same time as you are upskilling people. 669 00:33:08,616 --> 00:33:11,592 The great benefit of a learning hour, of course, is you can take 670 00:33:11,592 --> 00:33:14,738 a skill that you need for your production code and you can 671 00:33:14,738 --> 00:33:18,240 practice it on a simpler environment where you can repeat 672 00:33:18,240 --> 00:33:22,550 it and really start to get that fluency with it. 673 00:33:22,790 --> 00:33:25,740 And to make sure you remember all the steps so that when you 674 00:33:25,740 --> 00:33:29,162 meet that same situation in your production code, that it comes 675 00:33:29,162 --> 00:33:32,899 naturally to you. So I think for a healthy team, 676 00:33:32,899 --> 00:33:35,915 they'll have some learning hours, but they'll also have 677 00:33:35,915 --> 00:33:39,511 some collaborative coding, some pairing and ensemble in order to 678 00:33:39,511 --> 00:33:41,827 level up the more junior team members. 679 00:33:42,327 --> 00:33:45,857 And also the senior members, I mean, they benefit from this 680 00:33:45,857 --> 00:33:49,535 too. So if you are sitting with a 681 00:33:49,535 --> 00:33:52,803 junior and they're not getting something, you have to explain 682 00:33:52,803 --> 00:33:55,797 it to them. You have to learn to put into 683 00:33:55,797 --> 00:33:59,561 words what it is you do and to explain why you do that. 684 00:33:59,801 --> 00:34:02,681 And just that kind of being forced to explain helps you to 685 00:34:02,681 --> 00:34:06,077 understand it better yourself. So I think it really benefits 686 00:34:06,077 --> 00:34:09,735 the senior team members to also level up their skills by having 687 00:34:09,735 --> 00:34:11,967 those juniors that they need to explain stuff to. 688 00:34:12,617 --> 00:34:16,110 So I think this is one of the sort of the great strengths of 689 00:34:16,110 --> 00:34:19,130 the way we organize software developers today is into teams 690 00:34:19,130 --> 00:34:22,989 with a mix of, you know, years of experience that you've got 691 00:34:22,989 --> 00:34:25,880 juniors and seniors who have this vested interest in, well, 692 00:34:25,880 --> 00:34:27,679 we're gonna work together, we're a team. 693 00:34:28,039 --> 00:34:31,849 If that other person upskills, then it makes my life easier. 694 00:34:32,149 --> 00:34:34,938 So I have an interest in coaching them and mentoring 695 00:34:34,938 --> 00:34:37,176 them. And yeah, so that's the way a 696 00:34:37,176 --> 00:34:39,132 healthy team should be from my perspective. 697 00:34:40,005 --> 00:34:42,188 Yeah, so I like that you mentioned about pull requests, 698 00:34:42,188 --> 00:34:43,833 right? I think maybe in many software 699 00:34:43,833 --> 00:34:46,288 engineering team, this is like the workflow that they adopt, 700 00:34:46,288 --> 00:34:48,436 right? So like a pull request being 701 00:34:48,436 --> 00:34:51,577 raised and somebody review. Good thing if the pull request 702 00:34:51,577 --> 00:34:53,958 is small. But if it's large, I guess it's 703 00:34:53,958 --> 00:34:55,547 very difficult to actually understand, even. 704 00:34:56,031 --> 00:34:58,546 And sometimes like the challenge is like what you mentioned is an 705 00:34:58,546 --> 00:35:01,156 after effect thing, right? So the code has already been 706 00:35:01,156 --> 00:35:03,574 written, the thought process is missing, the context probably is 707 00:35:03,574 --> 00:35:05,332 missing. And you, yourself has your own 708 00:35:05,332 --> 00:35:07,952 thing that you work on that maybe still lingering in your 709 00:35:07,952 --> 00:35:09,708 mind and you just couldn't forget, right? 710 00:35:10,128 --> 00:35:12,801 And yeah, it's very difficult to actually learn from such 711 00:35:12,801 --> 00:35:15,340 situation. And I guess the pull request 712 00:35:15,340 --> 00:35:19,206 thing can be detrimental to the learning that people can have, 713 00:35:19,206 --> 00:35:21,257 right, so compared to like pair programming. 714 00:35:21,580 --> 00:35:24,080 Yeah, I wouldn't say it was detrimental. 715 00:35:24,080 --> 00:35:28,240 I mean, I think there's a lot of evidence that code review is a 716 00:35:28,300 --> 00:35:32,011 useful practice, so I don't wanna tell people not to do code 717 00:35:32,011 --> 00:35:34,494 review. But I just think you can't pin 718 00:35:34,494 --> 00:35:38,210 all your hopes on it, that this is gonna do all of the 719 00:35:38,210 --> 00:35:41,198 mentoring, you know. This is not something the, 720 00:35:41,198 --> 00:35:44,143 shouldn't be the primary mechanism to give advice to 721 00:35:44,143 --> 00:35:46,758 junior developers. So I'm not sure that it, yeah, 722 00:35:46,758 --> 00:35:48,286 I'm not trying to say don't do it. 723 00:35:48,406 --> 00:35:52,106 Although I happen to be a big fan of trunk-based development. 724 00:35:52,554 --> 00:35:55,464 I think that's not a battle I really wanna fight. 725 00:35:55,860 --> 00:35:59,320 The thing I'm most interested in is making sure that senior 726 00:35:59,320 --> 00:36:01,936 developers actually collaborate with the rest of the team 727 00:36:01,936 --> 00:36:05,732 members, and don't just think that it's enough to write a pull 728 00:36:05,732 --> 00:36:07,178 request comment from time to time. 729 00:36:07,283 --> 00:36:09,928 That's not being a senior developer, you know? 730 00:36:10,695 --> 00:36:12,675 Yeah, I was about to mention about comments as well, right? 731 00:36:12,675 --> 00:36:14,445 Because in a pull request, typically. 732 00:36:14,745 --> 00:36:17,415 If you put a lot of comments, the longer it will take, because 733 00:36:17,415 --> 00:36:20,252 it's async as well, and people kind of like multitasking, 734 00:36:20,252 --> 00:36:22,540 right? So obviously maybe trunk based 735 00:36:22,540 --> 00:36:25,166 development is something that people should aspire to. 736 00:36:25,286 --> 00:36:28,226 But yeah, definitely different team has different contexts and 737 00:36:28,226 --> 00:36:30,022 the practice that works best for them. 738 00:36:30,562 --> 00:36:32,002 So one thing about learning, right? 739 00:36:32,002 --> 00:36:35,404 So I think it's ideal, every team should have this learning 740 00:36:35,404 --> 00:36:38,014 hours probably. But you know, these days there's 741 00:36:38,014 --> 00:36:41,072 so many software engineering team that are just, you know, 742 00:36:41,072 --> 00:36:43,628 bogged down with so many different tasks, right? 743 00:36:44,018 --> 00:36:46,985 Uh, and so many different things that they have to do with such a 744 00:36:46,985 --> 00:36:49,820 small team, right? So again, like how do you 745 00:36:49,820 --> 00:36:52,889 actually do the elevator pitch, like as a team, right? 746 00:36:52,889 --> 00:36:56,137 So maybe once a sprint you can spend, I don't know, quality 747 00:36:56,137 --> 00:36:57,667 time to actually do some learning. 748 00:36:57,907 --> 00:37:00,487 Because some people are really struggling to find the time to 749 00:37:00,517 --> 00:37:05,086 actually do this kind of thing. Yeah, I know it's, that's one of 750 00:37:05,086 --> 00:37:08,767 the great benefits of being a technical coach is that I'm not 751 00:37:08,767 --> 00:37:12,479 responsible for delivering any of those tickets, so I can 752 00:37:12,479 --> 00:37:16,100 really, uh, have that time to prepare training materials. 753 00:37:16,515 --> 00:37:21,747 And when I come into a team, I can help them without needing to 754 00:37:21,747 --> 00:37:23,903 get involved with the deadlines that they're facing. 755 00:37:23,963 --> 00:37:26,313 So that's a great, you know, privilege I have. 756 00:37:26,953 --> 00:37:30,250 But yes, if you are a tech lead and a team with delivery 757 00:37:30,250 --> 00:37:32,537 responsibility, it's hard to think, oh, I've gotta organize a 758 00:37:32,537 --> 00:37:35,664 learning hour as well. And that's one of the things 759 00:37:35,664 --> 00:37:39,372 that I'm, I think for it to really, you know, for this to 760 00:37:39,372 --> 00:37:41,947 happen, people to be upskilled, that it's not enough that 761 00:37:41,947 --> 00:37:44,232 there's a bunch of people like me running around being 762 00:37:44,232 --> 00:37:46,745 technical coaches. We need to make it possible also 763 00:37:46,745 --> 00:37:48,713 for normal team leads to do learning hours. 764 00:37:49,073 --> 00:37:52,793 So one of the things I've done is found a society, a 765 00:37:52,793 --> 00:37:55,023 not-for-profit organization, the Samman Society. 766 00:37:55,663 --> 00:37:59,603 And we try and provide resources for all technical coaches and 767 00:37:59,628 --> 00:38:03,063 all software developers to make it easier to learn these things. 768 00:38:03,363 --> 00:38:06,690 So we've got a website and we have online meetings where you 769 00:38:06,690 --> 00:38:10,121 can meet other coaches. But published on the website, we 770 00:38:10,121 --> 00:38:13,212 publish code katas. We publish also descriptions of 771 00:38:13,212 --> 00:38:16,421 learning hours. So you can go on there and, you 772 00:38:16,421 --> 00:38:19,255 know, pick a topic that you, you'd like your team to upskill 773 00:38:19,255 --> 00:38:21,498 in. There you find hopefully some 774 00:38:21,498 --> 00:38:24,024 materials, some suggestions for how to structure a session. 775 00:38:24,355 --> 00:38:28,613 Some code katas to work on. Hopefully that's gonna make it 776 00:38:28,613 --> 00:38:31,378 easier for normal developers to do this. 777 00:38:32,098 --> 00:38:34,815 And then of course, I've also got the whole, uh, YouTube 778 00:38:34,815 --> 00:38:37,188 channel thing. Written descriptions of a 779 00:38:37,188 --> 00:38:40,924 learning hour is great, but I find video is a great medium for 780 00:38:40,924 --> 00:38:45,014 showing coding techniques. So a lot of the videos that I 781 00:38:45,014 --> 00:38:47,070 produce are little demos of refactoring techniques and 782 00:38:47,070 --> 00:38:50,292 little code katas and stuff that you could try out as well. 783 00:38:50,872 --> 00:38:52,712 So I'm doing more of that as well these days. 784 00:38:53,482 --> 00:38:55,295 Wow! That I think is really helpful, 785 00:38:55,295 --> 00:38:56,916 right? So for people, again, many 786 00:38:56,916 --> 00:38:59,636 people might be new to these kind of techniques, right? 787 00:38:59,936 --> 00:39:02,126 Having resources definitely is gonna help them. 788 00:39:02,186 --> 00:39:04,621 And I think I've seen a couple of your videos on YouTube. 789 00:39:04,946 --> 00:39:08,426 They're really interactive and you know, you can see exactly, 790 00:39:08,555 --> 00:39:10,980 you know, how you do stuff. I think that's definitely a 791 00:39:10,980 --> 00:39:13,810 thing that people should, you know, maybe learn something from 792 00:39:13,810 --> 00:39:17,244 your resources. And for people who aspire to be 793 00:39:17,244 --> 00:39:20,042 technical coach, right? Because, I don't know, some 794 00:39:20,042 --> 00:39:24,333 people are just born into this, you know, engineering practices, 795 00:39:24,333 --> 00:39:27,100 TDD. They wanna upskill people doing 796 00:39:27,100 --> 00:39:30,627 quality stuff with their code. Is there a thing that you want 797 00:39:30,627 --> 00:39:33,325 to convey to them? How can they become a better 798 00:39:33,325 --> 00:39:37,279 technical coach? Yeah, so I think the world needs 799 00:39:37,279 --> 00:39:39,133 more technical coaches and technical coaching. 800 00:39:39,613 --> 00:39:42,490 And I think that it, as I said, if you, if you're a developer or 801 00:39:42,490 --> 00:39:45,683 a tech lead on a team and you start doing a few learning 802 00:39:45,683 --> 00:39:48,585 hours, then basically you are doing some technical coaching 803 00:39:48,585 --> 00:39:51,409 part-time. And if you are in an 804 00:39:51,409 --> 00:39:53,874 understanding organization who realizes the value of that, 805 00:39:53,874 --> 00:39:57,527 maybe they'll let you grow that role from being just a thing you 806 00:39:57,527 --> 00:40:00,433 do with your team from time to time that maybe you can lead 807 00:40:00,433 --> 00:40:03,442 learning hours with other teams. And maybe you could start doing 808 00:40:03,442 --> 00:40:06,989 some ensemble and, you know, start to build it up as a bigger 809 00:40:06,989 --> 00:40:09,783 part of your role. And maybe even then transfer 810 00:40:09,783 --> 00:40:13,212 into a, like a staff engineer role in your organization where 811 00:40:13,212 --> 00:40:16,513 you've got responsibilities for more than one team. 812 00:40:16,543 --> 00:40:19,230 And part of your role could be doing these kind of technical 813 00:40:19,230 --> 00:40:23,007 coaching activities alongside, I mean, staff engineers do loads 814 00:40:23,007 --> 00:40:26,173 of variety of important technical leadership things. 815 00:40:26,473 --> 00:40:29,473 And I think you can combine it with some technical coaching. 816 00:40:30,133 --> 00:40:33,147 And then, uh, maybe if you really like this side of the 817 00:40:33,147 --> 00:40:36,534 work, you could go full-time like I have and be an 818 00:40:36,534 --> 00:40:40,488 independent coach. But I kind of see the, the path 819 00:40:40,488 --> 00:40:43,535 into this through basically a part-time role alongside tech 820 00:40:43,535 --> 00:40:45,796 lead and growing that in your organization. 821 00:40:46,986 --> 00:40:50,604 Although, I mean there are organizations that are hiring 822 00:40:50,604 --> 00:40:53,942 technical coaches and maybe be lucky enough to, to get a 823 00:40:53,942 --> 00:40:57,177 full-time role straight away, but I think it's more likely it 824 00:40:57,177 --> 00:40:59,307 would be alongside a staff engineer role. 825 00:41:00,165 --> 00:41:01,955 Yeah. So I think, I mean definitely 826 00:41:01,955 --> 00:41:04,472 there are people who are interested in doing this, right, 827 00:41:04,472 --> 00:41:07,302 within the team. I think the message here is just 828 00:41:07,302 --> 00:41:10,652 keep continue doing that, right? And that's also improving your 829 00:41:10,652 --> 00:41:12,932 skills, right? Maybe your soft skills, maybe 830 00:41:12,932 --> 00:41:15,482 your teaching skills, right? And also learning new things, 831 00:41:15,482 --> 00:41:17,397 right? Because I'm sure you want to 832 00:41:17,397 --> 00:41:20,445 learn new things together with others as well and making sure 833 00:41:20,445 --> 00:41:22,977 that people also learn benefit from the learning, right? 834 00:41:23,397 --> 00:41:27,004 So I think you mentioned about being technical coach. 835 00:41:27,094 --> 00:41:28,744 How do you actually get measured? 836 00:41:28,774 --> 00:41:31,354 This is something that I'm quite curious because is it more like 837 00:41:31,354 --> 00:41:35,657 an engagement, uh, with a set of, I don't know, teachable 838 00:41:35,657 --> 00:41:38,088 skills? Or is there something else that 839 00:41:38,088 --> 00:41:41,695 you measure quantifiable? Yeah, this is one of the hardest 840 00:41:41,695 --> 00:41:44,805 parts of my job 'cause it really just does not interest me at 841 00:41:44,805 --> 00:41:48,587 all. This whole thing of metrics and 842 00:41:48,587 --> 00:41:52,816 performance and, yeah, that's managers obviously do a lot more 843 00:41:52,816 --> 00:41:54,722 of that. And I'm, I've made a deliberate 844 00:41:54,722 --> 00:41:56,612 choice in my career not to be a manager. 845 00:41:57,103 --> 00:41:59,496 All of this kind of like, how do we measure if the technical 846 00:41:59,496 --> 00:42:02,439 coach is effective? My usual answer is, well, ask 847 00:42:02,439 --> 00:42:05,037 the team. What did they think this has 848 00:42:05,037 --> 00:42:08,058 given them? And try and perhaps do a survey 849 00:42:08,058 --> 00:42:10,721 before and after the coaching about their attitudes towards 850 00:42:10,721 --> 00:42:15,015 teamwork and, and ensemble and ask about their attitudes to TDD 851 00:42:15,015 --> 00:42:18,930 and code quality, and see if there's a shift through the 852 00:42:18,930 --> 00:42:21,628 coaching. And then also, while the 853 00:42:21,628 --> 00:42:25,128 coaching is going on, I like to write little summaries that I 854 00:42:25,128 --> 00:42:27,919 send to the managers about what happens during that coaching 855 00:42:27,919 --> 00:42:30,173 session. And really just like a 856 00:42:30,173 --> 00:42:32,872 paragraph. No more than two or three 857 00:42:32,872 --> 00:42:36,238 sentences really just to describe the what happened and 858 00:42:37,033 --> 00:42:41,453 the impact that I think it should be having on the team and 859 00:42:41,453 --> 00:42:43,913 the organization. So the manager gets then, at the 860 00:42:43,913 --> 00:42:46,582 end of the coaching, summary of all of those little summaries, 861 00:42:46,582 --> 00:42:50,113 it's a bit meta. They get a kind of an idea 862 00:42:50,113 --> 00:42:53,274 about, well, this is what I, as the coach, think that the team 863 00:42:53,274 --> 00:42:55,930 has got outta this coaching. And then they can go and 864 00:42:55,930 --> 00:42:58,359 basically follow up. Do they observe that continuing 865 00:42:58,359 --> 00:43:02,719 after the coach left? That's the way I usually try and 866 00:43:02,719 --> 00:43:04,773 help people to understand what is the value of this. 867 00:43:05,952 --> 00:43:09,007 Yeah, it's always difficult to kind of like quantify coaching, 868 00:43:09,007 --> 00:43:11,235 right? Because sometimes it's like more 869 00:43:11,235 --> 00:43:13,301 abstract, right? There's not immediate impact as 870 00:43:13,301 --> 00:43:15,655 well, right? It's not like suddenly you can 871 00:43:15,655 --> 00:43:18,397 do code faster, for example, produce more lines of code. 872 00:43:18,427 --> 00:43:22,083 Um, but I think... Yeah, you hope, you hope that 873 00:43:22,083 --> 00:43:25,370 people will be more productive and better coders afterwards, 874 00:43:25,370 --> 00:43:28,660 but those things are really hard to measure and they... 875 00:43:29,240 --> 00:43:33,491 It might take a while before you see, you know, an increase in 876 00:43:33,491 --> 00:43:35,268 productivity, if the code quality is very low. 877 00:43:35,268 --> 00:43:39,608 It could take a while to raise it, you know, so I think there 878 00:43:39,608 --> 00:43:43,148 are people much more interested in those questions than I am, 879 00:43:43,148 --> 00:43:46,296 who would be able to tell you more about how to measure 880 00:43:46,296 --> 00:43:47,900 developer productivity and the impact of some coaching. 881 00:43:48,180 --> 00:43:50,860 So I just gonna dodge the question, basically. 882 00:43:52,857 --> 00:43:55,618 So, let's go back maybe to the AI case, right? 883 00:43:55,618 --> 00:43:59,386 So I'm sure with the advancement of AI these days, like 884 00:43:59,386 --> 00:44:02,475 especially the agentic thing that could happen, you know, 885 00:44:02,475 --> 00:44:05,182 asynchronously. And there's a danger that people 886 00:44:05,182 --> 00:44:07,734 probably do a lot of multitasking, accepting AI 887 00:44:07,734 --> 00:44:11,270 suggestions or even just merging the code straight away, right, 888 00:44:11,270 --> 00:44:14,346 to production. As a technical coach, maybe you 889 00:44:14,346 --> 00:44:16,296 have something to say about this. 890 00:44:16,296 --> 00:44:18,491 What do you think? Like your prediction, like what 891 00:44:18,491 --> 00:44:20,221 could happen if this trend continues? 892 00:44:21,539 --> 00:44:23,239 Yeah. So I've definitely heard that, 893 00:44:23,239 --> 00:44:26,367 that people are, you know, vibe coding and that can mean not 894 00:44:26,367 --> 00:44:28,622 looking at the code the thing produces at all. 895 00:44:29,162 --> 00:44:32,178 Just kind of accepting, oh, it looks, it looks good to me, 896 00:44:32,178 --> 00:44:36,166 let's just ship it. And of course, I see a danger in 897 00:44:36,166 --> 00:44:38,917 the code quality there, not being quite what you'd like it 898 00:44:38,917 --> 00:44:41,402 to be. And I kind of have this notion 899 00:44:41,402 --> 00:44:43,758 that as a software developer, I'm responsible for the quality 900 00:44:43,758 --> 00:44:47,164 of the work that I deliver. And if I haven't even looked at 901 00:44:47,164 --> 00:44:50,815 the code, I mean, how do I know that it's gonna be even vaguely 902 00:44:50,815 --> 00:44:54,663 something I could stand up for? I mean, okay. 903 00:44:54,723 --> 00:44:58,082 Okay. I'm, I want to use these tools 904 00:44:58,082 --> 00:45:03,067 to augment my abilities and to write even better code than I'd 905 00:45:03,067 --> 00:45:06,964 be capable of without the tool. So, uh, I want to look at the 906 00:45:06,964 --> 00:45:08,672 code that it produces before I ship it. 907 00:45:08,702 --> 00:45:12,536 I think that's actually quite fundamental, that a software 908 00:45:12,536 --> 00:45:15,682 developer should be responsible for the code they produce. 909 00:45:16,618 --> 00:45:18,163 Yeah. But one challenge with these 910 00:45:18,163 --> 00:45:20,109 tools is like kind of seductive, right? 911 00:45:20,109 --> 00:45:23,070 Like it can churn out code really, really fast 912 00:45:23,070 --> 00:45:24,819 automatically. And I've experienced myself 913 00:45:24,819 --> 00:45:28,578 using some newer models, they're like hungry to give you like a 914 00:45:28,578 --> 00:45:30,974 lot of output, right? You know, like maybe I could do 915 00:45:30,974 --> 00:45:33,944 this, I could do that and keep on doing stuff that sometimes 916 00:45:33,944 --> 00:45:37,238 yeah, you wanna see how far it goes, right? 917 00:45:37,294 --> 00:45:39,467 But after they have done it, right? 918 00:45:39,486 --> 00:45:42,242 It's like so many things. I think it's really hard to 919 00:45:42,242 --> 00:45:43,808 review. Maybe I don't know whether you 920 00:45:43,808 --> 00:45:45,338 have this special case that you have seen before? 921 00:45:46,066 --> 00:45:48,882 Maybe I need to qualify this with, if you're writing a 922 00:45:48,882 --> 00:45:52,413 prototype or proof of concept to show that you could write 923 00:45:52,413 --> 00:45:56,364 software to solve this problem or... and, uh, it's not gonna go 924 00:45:56,364 --> 00:45:59,418 into production and... then that's a very good use case for 925 00:45:59,418 --> 00:46:01,742 just doing, as you say, just get it to produce a huge amount of 926 00:46:01,742 --> 00:46:04,539 code that just demonstrate what could happen. 927 00:46:05,109 --> 00:46:08,699 But I think if most of the teams I'm working with are not in that 928 00:46:08,699 --> 00:46:11,873 situation, they're, they've got a large existing code base, 929 00:46:11,873 --> 00:46:15,664 which is pretty complex and valuable and is producing a lot 930 00:46:15,664 --> 00:46:20,355 of money for the company. And if you are using the AI to 931 00:46:20,355 --> 00:46:24,788 produce a load of code that is gonna be unmaintainable and 932 00:46:24,788 --> 00:46:30,433 integrate with this existing code base, it's a bit of a 933 00:46:30,433 --> 00:46:33,002 recipe for disaster. I think it's gonna come back and 934 00:46:33,002 --> 00:46:35,404 bite you pretty soon. So most of the developers I'm 935 00:46:35,404 --> 00:46:38,518 working with, I'm saying, look, yeah, use the AI tool to get 936 00:46:38,518 --> 00:46:41,996 ideas and suggestions, but you've gotta check it and you've 937 00:46:41,996 --> 00:46:45,513 gotta be making sure that the changes that it's making are in 938 00:46:45,513 --> 00:46:48,649 line with the rest of the system that's already there and that 939 00:46:48,649 --> 00:46:50,385 you're gonna be able to maintain it. 940 00:46:50,385 --> 00:46:54,152 So I think, yeah, I advise against just producing huge 941 00:46:54,152 --> 00:46:56,100 amounts of code that you don't really understand. 942 00:46:56,100 --> 00:46:58,286 That, that doesn't seem like a recipe for success. 943 00:46:59,393 --> 00:47:01,993 Right. And this gives a much more 944 00:47:01,993 --> 00:47:05,141 importance for, I dunno, reading code, rationalizing the design 945 00:47:05,141 --> 00:47:07,553 that is, you know, generated, right? 946 00:47:07,853 --> 00:47:10,834 So do you think, uh, there are like more prominent new skills 947 00:47:10,834 --> 00:47:12,934 that, I don't know, like technical people or maybe 948 00:47:12,934 --> 00:47:14,513 technical coach needs to be aware of? 949 00:47:14,900 --> 00:47:16,826 That something, you know, because of AI, right? 950 00:47:17,111 --> 00:47:19,151 We have to be more careful about this. 951 00:47:19,151 --> 00:47:22,424 Or maybe some skills, I dunno, like reading code, making sure 952 00:47:22,424 --> 00:47:25,781 the code is well written in terms of variable naming, design 953 00:47:25,781 --> 00:47:28,159 and all that. Maybe some things that you can 954 00:47:28,159 --> 00:47:31,594 advise us as well. Yeah, I'm not, I think the 955 00:47:31,594 --> 00:47:34,591 fundamentals are still the same. The modern software engineering 956 00:47:34,591 --> 00:47:39,544 skills of being able to look at a problem and break down the 957 00:47:39,544 --> 00:47:42,715 complexity of it into a modular design. 958 00:47:43,325 --> 00:47:47,404 And being able to work out how to build that design in small 959 00:47:47,404 --> 00:47:50,246 steps and react. As you get more code, you 960 00:47:50,246 --> 00:47:52,564 realize that the design you initially had in your head maybe 961 00:47:52,564 --> 00:47:54,948 wasn't the best. And you can adapt and move the 962 00:47:54,948 --> 00:47:58,480 design in a different direction. So it's about being able to 963 00:47:58,480 --> 00:48:02,664 explain to the AI tool this is the problem I want you to solve 964 00:48:02,664 --> 00:48:06,456 now concisely and then to verify that the code it produces does 965 00:48:06,456 --> 00:48:08,520 that. So as you say, code reading 966 00:48:08,520 --> 00:48:10,491 skills. And then refactoring skills to 967 00:48:10,491 --> 00:48:12,981 say, actually I wanna move this in a different direction. 968 00:48:13,636 --> 00:48:17,193 At the time I'm speaking now, the AI tools aren't that good at 969 00:48:17,193 --> 00:48:19,916 refactoring. Um, they tend to introduce 970 00:48:19,916 --> 00:48:24,458 errors and do unexpected things. I still think it's a really 971 00:48:24,458 --> 00:48:27,420 useful to learn refactoring tools, deterministic refactoring 972 00:48:27,420 --> 00:48:31,481 tools, because they do it in a reliable way that is 973 00:48:31,481 --> 00:48:34,028 predictable. I think that, uh, the 974 00:48:34,028 --> 00:48:35,610 refactoring skills are still valuable. 975 00:48:35,850 --> 00:48:38,472 But the skill I, you know, describing this is the piece of 976 00:48:38,472 --> 00:48:41,102 the problem I wanna solve now, that's test-driven development. 977 00:48:41,122 --> 00:48:45,354 That's being able to break down a problem and describe what the 978 00:48:45,354 --> 00:48:48,879 next bit should do. The abstraction level has gone 979 00:48:48,879 --> 00:48:52,401 up a little, and I can describe it more in words than in a test 980 00:48:52,401 --> 00:48:53,806 case. But still, I think it's 981 00:48:53,806 --> 00:48:56,876 fundamentally the same skills that you need to be a good 982 00:48:56,876 --> 00:49:00,706 engineer today as it were before AI was invented, actually. 983 00:49:02,683 --> 00:49:04,753 Yeah, I like that you mentioned about refactoring, right? 984 00:49:04,753 --> 00:49:07,689 I, from my experience as well, using these AI tools, I feel 985 00:49:07,689 --> 00:49:09,973 that refactoring is very, very crucial. 986 00:49:10,033 --> 00:49:12,373 And if we think about the TDD loop, right? 987 00:49:12,403 --> 00:49:14,653 You know, red, green, refactor, right? 988 00:49:14,983 --> 00:49:19,590 I think these days like prompt, I dunno, uh, red, green, the 989 00:49:19,590 --> 00:49:21,553 refactoring part is actually still very important. 990 00:49:21,553 --> 00:49:24,513 Because I can see AI generating code that is still kind of like 991 00:49:24,513 --> 00:49:27,364 procedural, long lines of code. And even they just put comments 992 00:49:27,364 --> 00:49:30,505 on top of some sections of the code rather than creating, you 993 00:49:30,505 --> 00:49:34,310 know, methods. So I think, um, not, not, uh, 994 00:49:34,310 --> 00:49:37,899 doing refactoring is I think a big mistake, uh, especially if 995 00:49:37,899 --> 00:49:39,914 the software needs to be maintainable in the future, 996 00:49:39,914 --> 00:49:41,601 right? 'Cause you could end up like 997 00:49:41,601 --> 00:49:44,502 with so many lines of code in just one particular class or 998 00:49:44,502 --> 00:49:47,192 function, right? And it gets so difficult to 999 00:49:47,192 --> 00:49:50,088 understand over the time. So I think thanks for mentioning 1000 00:49:50,088 --> 00:49:52,906 about refactoring. I think for people who are still 1001 00:49:52,906 --> 00:49:55,345 not familiar with the refactoring technique, please do 1002 00:49:55,345 --> 00:49:58,420 learn about that. And what I'm seeing quite 1003 00:49:58,420 --> 00:50:01,801 interesting as well, people actually do the refactoring by 1004 00:50:01,801 --> 00:50:03,139 prompting the AI to do the refactoring. 1005 00:50:04,129 --> 00:50:06,259 Uh. Although it's not deterministic, 1006 00:50:06,259 --> 00:50:08,025 right? There's no like compiler thing 1007 00:50:08,025 --> 00:50:11,205 and the structure of the code, static code analysis that is 1008 00:50:11,205 --> 00:50:13,997 being used. Sometimes it gets to produce a 1009 00:50:13,997 --> 00:50:16,604 decent result. But what do you think about this 1010 00:50:16,604 --> 00:50:18,493 refactoring through AI? Because it's, again, like I 1011 00:50:18,493 --> 00:50:19,695 said, right, it's quite seductive. 1012 00:50:19,695 --> 00:50:22,149 You could just instruct something to do something for 1013 00:50:22,149 --> 00:50:25,053 you rather than, you know, the mechanics of, you know, doing 1014 00:50:25,053 --> 00:50:26,355 all this refactoring by yourself. 1015 00:50:27,564 --> 00:50:30,684 I think there's studies to show this, that it actually, AI tools 1016 00:50:30,684 --> 00:50:33,911 are very bad at refactoring. They make mistakes, they 1017 00:50:33,911 --> 00:50:37,583 introduce bugs and they forget things or they introduce dead 1018 00:50:37,583 --> 00:50:40,091 code or, you know, they do random stuff. 1019 00:50:40,181 --> 00:50:42,873 And I think the tools are bound to get better. 1020 00:50:42,873 --> 00:50:44,343 They seem to be getting better all the time. 1021 00:50:44,343 --> 00:50:47,547 But at the moment, I don't think you should trust it to do 1022 00:50:47,547 --> 00:50:49,383 refactoring, particularly, if you've got a significant amount 1023 00:50:49,383 --> 00:50:52,338 of code in your project. It won't find all the places to 1024 00:50:52,338 --> 00:50:53,905 update. It will do random things. 1025 00:50:54,325 --> 00:50:59,313 But we do have, for many years now, reliable deterministic 1026 00:50:59,313 --> 00:51:02,549 refactoring tools. Use them, for goodness sake! 1027 00:51:03,109 --> 00:51:05,535 I think it's much better to use a reliable tool that you know 1028 00:51:05,535 --> 00:51:08,039 what it's gonna do, and it's gonna find all the right spots 1029 00:51:08,039 --> 00:51:10,229 to update than to ask the AI to do it. 1030 00:51:10,419 --> 00:51:15,679 And the cost, of course, is that you need to learn how to use the 1031 00:51:15,679 --> 00:51:18,729 refactoring tool and learn what the thing is called and where, 1032 00:51:18,729 --> 00:51:21,369 how to prompt, you know, where to put the cursor. 1033 00:51:21,657 --> 00:51:24,139 One thing I really realize now, I've been teaching refactoring 1034 00:51:24,139 --> 00:51:28,029 tools to so many developers, how un-user friendly these tools 1035 00:51:28,029 --> 00:51:31,244 really are. I mean, you have to get the 1036 00:51:31,244 --> 00:51:34,424 cursor in exactly the right spots for the menu item to even 1037 00:51:34,424 --> 00:51:38,059 appear. And then the name of the menu 1038 00:51:38,059 --> 00:51:41,667 item is something that isn't perhaps anything you recognize. 1039 00:51:41,998 --> 00:51:46,290 It's called transform parameter or introduce parameter object. 1040 00:51:46,710 --> 00:51:49,024 And when I first read those, I was like, well, I dunno what 1041 00:51:49,024 --> 00:51:50,959 that is. I'm a bit scared to click on 1042 00:51:50,959 --> 00:51:52,230 that. I don't know what it's gonna do. 1043 00:51:52,710 --> 00:51:56,236 Um, and then you... but so you, that is the thing that, you 1044 00:51:56,236 --> 00:51:59,144 know, that the usability isn't there, whereas the AI tool, you 1045 00:51:59,144 --> 00:52:03,171 can say, okay, I need to change the class of that parameter and, 1046 00:52:03,171 --> 00:52:06,649 uh, introduce an adapter pattern and it might understand you. 1047 00:52:07,789 --> 00:52:10,549 So I think that the, uh, refactoring, deterministic 1048 00:52:10,549 --> 00:52:12,859 refactoring tools have got a usability problem. 1049 00:52:12,859 --> 00:52:16,644 But I think the solution is not to just tell people to use AI 1050 00:52:16,644 --> 00:52:18,427 instead. Right. 1051 00:52:18,427 --> 00:52:20,287 So you mentioned a very valid point, right? 1052 00:52:20,287 --> 00:52:23,167 So I think the vocabulary, right, the techniques. 1053 00:52:23,227 --> 00:52:25,717 I think Martin Fowler has a book about refactoring. 1054 00:52:25,717 --> 00:52:28,605 I think some other people, maybe like Michael Feathers' Working 1055 00:52:28,605 --> 00:52:30,893 with Legacy Code has some techniques for doing this 1056 00:52:30,893 --> 00:52:33,355 refactoring kind of thing. Definitely, knowing the 1057 00:52:33,355 --> 00:52:36,392 vocabulary helps, right? But getting to master the tools 1058 00:52:36,392 --> 00:52:38,972 in order to do that deterministically, I think it's 1059 00:52:38,972 --> 00:52:40,892 also still crucial for software developers as well. 1060 00:52:41,552 --> 00:52:44,756 So I think it's been a great thing to learn from you, Emily. 1061 00:52:44,756 --> 00:52:48,240 So I dunno whether there are a thing or two about ensemble or 1062 00:52:48,240 --> 00:52:50,396 Samman method that you still wanna convey to us. 1063 00:52:50,396 --> 00:52:52,286 Maybe is there a thing that you wanna plug in? 1064 00:52:53,275 --> 00:52:55,283 I would encourage anyone interested in technical coaching 1065 00:52:55,283 --> 00:52:58,753 or the Samman method to come and look at our website, 1066 00:52:58,753 --> 00:53:00,589 sammancoaching.org. Uh, we have a newsletter that 1067 00:53:00,589 --> 00:53:03,388 you can sign up for and then you'll hear about updates to the 1068 00:53:03,388 --> 00:53:05,294 website and new materials that we publish. 1069 00:53:05,904 --> 00:53:08,793 And of course, we have these in-person shorts, online 1070 00:53:08,793 --> 00:53:11,706 meetings every month. If you're on a suitable time 1071 00:53:11,706 --> 00:53:14,612 zone to come along. You can come and meet me and 1072 00:53:14,612 --> 00:53:17,406 other coaches at those. So that's something to try if 1073 00:53:17,406 --> 00:53:20,473 you're interested. Yeah, sounds like it's gonna be 1074 00:53:20,473 --> 00:53:22,411 fun, right? So having technical coaches 1075 00:53:22,411 --> 00:53:24,949 together and exchanging, you know, good practices. 1076 00:53:25,249 --> 00:53:28,631 I think definitely it'll amplify a lot of the skills, right? 1077 00:53:29,201 --> 00:53:32,321 So Emily, I only have one last question before, uh, we wrap up 1078 00:53:32,321 --> 00:53:34,654 this conversation. I normally, call this the three 1079 00:53:34,654 --> 00:53:37,108 technical leadership wisdom. I always ask this to my guests. 1080 00:53:37,108 --> 00:53:40,074 If you can think of it just like advice that you want to give to 1081 00:53:40,074 --> 00:53:41,933 the listeners. Maybe you can share your version 1082 00:53:41,933 --> 00:53:44,478 today, that would be great. Okay, okay. 1083 00:53:44,478 --> 00:53:47,209 So I've, I, you've told me this in advance, so I've prepared, 1084 00:53:47,209 --> 00:53:50,616 uh, three things. So the first technical 1085 00:53:50,616 --> 00:53:54,006 leadership wisdom is that junior developers need mentoring to 1086 00:53:54,006 --> 00:53:58,542 turn into senior developers. And mentoring is more than just 1087 00:53:58,542 --> 00:54:00,246 writing comments and pull requests. 1088 00:54:00,946 --> 00:54:03,806 Find a way to make sure you're actually mentoring your juniors. 1089 00:54:04,727 --> 00:54:07,452 The other thing is that about technical leadership is you 1090 00:54:07,452 --> 00:54:11,180 don't learn how to produce beautiful code only by looking 1091 00:54:11,180 --> 00:54:14,255 at a good design and analyzing it. 1092 00:54:14,615 --> 00:54:17,452 You learn to produce a good design by watching good 1093 00:54:17,452 --> 00:54:19,415 designers coming up with that design. 1094 00:54:19,805 --> 00:54:23,855 So you need to involve juniors in the design process and show 1095 00:54:23,855 --> 00:54:28,348 them not just the end result. Then the other thing I wanted to 1096 00:54:28,348 --> 00:54:31,051 say is that, in terms of technical leadership, that 1097 00:54:31,051 --> 00:54:33,376 sometimes it helps to have someone from outside like a 1098 00:54:33,376 --> 00:54:35,798 technical coach come in and help your team. 1099 00:54:36,516 --> 00:54:39,248 Our priority as a technical coach is to level up the team, 1100 00:54:39,248 --> 00:54:43,458 not to deliver features. And that could give you some 1101 00:54:43,458 --> 00:54:47,427 time and breathing space to actually focus on that and not 1102 00:54:47,427 --> 00:54:50,237 only delivering. So, uh, perhaps think about 1103 00:54:50,237 --> 00:54:54,503 getting in a technical coach for a short period or to level up if 1104 00:54:54,503 --> 00:54:57,548 you need that. Right, I think that's really 1105 00:54:57,548 --> 00:54:59,706 beautiful, right? So you emphasize a lot about 1106 00:54:59,706 --> 00:55:01,691 bringing people along into the journey, right? 1107 00:55:01,691 --> 00:55:04,494 The juniors or maybe the one who doesn't understand about good 1108 00:55:04,494 --> 00:55:06,538 design, right? Bringing them along to the 1109 00:55:06,538 --> 00:55:09,232 journey and making sure they learn from, you know, the 1110 00:55:09,232 --> 00:55:11,140 evolution, right? From like the bad to the good, 1111 00:55:11,140 --> 00:55:13,414 for example. But maybe for some people who 1112 00:55:13,414 --> 00:55:16,450 find it hard to find mentors, any last piece for them? 1113 00:55:16,480 --> 00:55:19,880 Like how do you actually find mentors that could actually 1114 00:55:19,880 --> 00:55:21,910 upskill you, bringing them along with the journey? 1115 00:55:22,908 --> 00:55:25,458 I spent a lot of my career looking for mentors. 1116 00:55:25,841 --> 00:55:29,238 I think it's really important to just, if you are in a team and 1117 00:55:29,238 --> 00:55:31,560 you're not getting that, then perhaps you need to look around 1118 00:55:31,560 --> 00:55:34,255 and see, is it, could you transfer to a team that has 1119 00:55:34,255 --> 00:55:37,784 somebody who will mentor you? And if not, then maybe I don't, 1120 00:55:37,784 --> 00:55:40,220 I hesitate to ask you to put your free time in. 1121 00:55:40,729 --> 00:55:44,297 But that is also an option. Go along to open ensembles and, 1122 00:55:44,297 --> 00:55:49,326 uh, events by the Samman Society and so on and see if you can get 1123 00:55:49,326 --> 00:55:52,738 people who will help you. I think everyone needs that to 1124 00:55:52,738 --> 00:55:54,478 progress in your career. You need mentors. 1125 00:55:55,630 --> 00:55:58,660 Yeah, so definitely go to the community, go to the forum, 1126 00:55:58,660 --> 00:56:01,740 maybe online, offline, right? Talk to people, I think that's 1127 00:56:01,740 --> 00:56:04,390 always still the best way to find good mentors, right? 1128 00:56:04,390 --> 00:56:06,920 Sometimes you are lucky you can find good mentors at work, 1129 00:56:06,920 --> 00:56:08,710 right? Working side by side with you. 1130 00:56:09,010 --> 00:56:10,930 I think that will be great definitely. 1131 00:56:10,930 --> 00:56:14,252 But for many people, sometimes this is not a thing that they 1132 00:56:14,252 --> 00:56:16,145 have, not fortunate enough to have that. 1133 00:56:16,505 --> 00:56:20,033 So definitely talk to people. Be active, talk around, uh, and 1134 00:56:20,033 --> 00:56:23,724 maybe join, uh, one of these sessions that Emily has. 1135 00:56:23,784 --> 00:56:26,000 I think that will be great things to find good mentors as 1136 00:56:26,000 --> 00:56:28,686 well. So thank you so much for 1137 00:56:28,686 --> 00:56:31,432 teaching us about Samman method, ensemble and importance of 1138 00:56:31,432 --> 00:56:34,129 learning hours. So I hope that people can do 1139 00:56:34,129 --> 00:56:37,543 practice of some of this. And also remember about the 1140 00:56:37,543 --> 00:56:40,868 danger using AI without actually caring about the quality or the 1141 00:56:40,868 --> 00:56:42,959 output that they produce. So thanks again, Emily. 1142 00:56:43,577 --> 00:56:44,927 Thank you so much for having me.