Transcriptly

"Don’t Learn to Code" Is WRONG | GitHub CEO Thomas Dohmke

EO19:07Added Aug 31, 2026

EO stands for Entrepreneurship & Opportunities. As we're looking to feature more inspiring stories of entrepreneurs all over the world, don't hesitate to con...

Watch on YouTube →
Contributed by 刘嘉琪

Transcript

Transcript format
  1. 00:00I first of all I strongly believe that
  2. 00:02every kid uh every child should learn
  3. 00:04coding. We should actually teach them
  4. 00:06coding in school in the same way that we
  5. 00:07teach them physics and geography and uh
  6. 00:10you know literacy and math and whatn
  7. 00:11not. Those are all fundamental skills.
  8. 00:13Coding is one such skill because
  9. 00:15software is everywhere and in fact our
  10. 00:17day-to-day is dominated by software
  11. 00:18already. You can't really live your life
  12. 00:20travel um wake up in the morning without
  13. 00:23software anymore. And I think as as
  14. 00:24humans it is crucial to not only be in
  15. 00:27readonly mode but also be able to create
  16. 00:30uh things ourselves or at least
  17. 00:31understand how how creation is done on
  18. 00:33these devices. My name is Thomas and I'm
  19. 00:34a developer. I've been developing uh
  20. 00:37software since the early 1990s and today
  21. 00:39my role is mostly being the GitHub CEO
  22. 00:42leading uh the largest developer
  23. 00:43platform uh on this planet.
  24. 00:48I was born in uh 1978 in East Berlin
  25. 00:51when Germany was divided into two
  26. 00:53countries. So I was on the on the
  27. 00:55eastern side uh in Berlin in in a suburb
  28. 00:57called Matsan. I grew you know had a
  29. 00:59normal childhood um but was fascinated
  30. 01:02by technology remote control cars uh uh
  31. 01:04you know little computer games. Uh in
  32. 01:061989 the wall fell and and that opened
  33. 01:09up you know a whole new world to me as
  34. 01:10an 11year-old. um obviously from from a
  35. 01:13toy perspective um getting access to to
  36. 01:15Lego and and Mickey Mouse but also uh to
  37. 01:18uh to computers and so in the early 90s
  38. 01:20um I bought my first computer uh 64 and
  39. 01:23after um university um I went to
  40. 01:26technical university in Berlin I
  41. 01:28actually uh went into the automotive
  42. 01:30industry uh I started working at uh
  43. 01:32Mercedes back then it was called Dameler
  44. 01:34Chrysler working on driver assistance
  45. 01:36system uh for for the S-Class and I did
  46. 01:38that uh for a while and then switched to
  47. 01:40the supplier side to Bosch uh uh which
  48. 01:43is you know a large automotive supplier
  49. 01:45amongst amongst other things and and
  50. 01:46there uh worked you know on parking
  51. 01:49systems and it was in 2008 uh that two
  52. 01:52things happened. One is I finished my
  53. 01:53PhD thesis but also Steve Jobs showed
  54. 01:56the iPhone SDK. Everybody wanted to
  55. 01:58build apps um and and at the time only
  56. 02:01Apple could build native apps and I
  57. 02:03thought this is this is cool. I got to,
  58. 02:05you know, build apps myself and I got to
  59. 02:06get into that space. And so I quit my
  60. 02:08job at Bosch uh at the height of the
  61. 02:10financial crisis amongst all things uh
  62. 02:12in late 2008 and um reconnected with a
  63. 02:15friend um from from university and um we
  64. 02:18just became you know two freelance
  65. 02:19developers building apps uh for the
  66. 02:22German market. So mostly agency work or
  67. 02:24or subcontractor for for larger app
  68. 02:26projects. I think in in 2009 and 10 we
  69. 02:29built around 30 apps. That work on
  70. 02:31building apps um for for for mostly
  71. 02:33German customers um German enterprise
  72. 02:35customers media and and automotive and
  73. 02:36so on got us the idea of of building a
  74. 02:39platform for mobile app developers. And
  75. 02:42so we um together with three friends
  76. 02:44Stefan Andreas and and Michael we
  77. 02:46founded a company called Hockey App
  78. 02:48which was a platform for mobile app
  79. 02:49developers. you could distribute your
  80. 02:51beta builds and um collect uh error
  81. 02:54reports, crash reports and feedback and
  82. 02:56those kind of things. And so we built
  83. 02:58that startup you know the platform for
  84. 03:00our own you know freelance business
  85. 03:02because we had that painoint that we
  86. 03:04wanted to solve uh for ourselves. Um
  87. 03:06because it was before that you would
  88. 03:08send you know a build uh over email you
  89. 03:10know to to a project manager um in
  90. 03:13Berlin or so and then the project
  91. 03:15manager would take that build and send
  92. 03:16it to the customer and explain to them
  93. 03:18or you have to you know drag and drop
  94. 03:20that into iTunes and connect you know uh
  95. 03:22uh the dock cable and and all these kind
  96. 03:24of things. since we made that way easier
  97. 03:26which made our own life easier and I
  98. 03:28think those are some of the greatest
  99. 03:30startup ideas when you're using it
  100. 03:31yourself uh uh day in day out and then
  101. 03:34improving it um based on the feedback
  102. 03:36that you're getting from your own
  103. 03:37customers. So we basically started
  104. 03:38building out the the hockey app business
  105. 03:40while we had the the contracting
  106. 03:42business and they were companies um both
  107. 03:45about the same size and um when
  108. 03:47Microsoft came in in 2014, we thought
  109. 03:50they would only buy the hockey air
  110. 03:52platform and the product business and
  111. 03:53and leave the contracting business alone
  112. 03:55uh given there were existing customer uh
  113. 03:58contracts and and what have you. uh but
  114. 04:00in fact Microsoft bought both companies
  115. 04:02uh the the subsidiary the product
  116. 04:04company and the the mothership the the
  117. 04:06contracting company because the
  118. 04:08contracting company had a number of iOS
  119. 04:10and Android developers uh that were a
  120. 04:12hot commodity uh in 2014 and and hard to
  121. 04:15hire um certainly you know at at startup
  122. 04:18rates and so Microsoft actually took
  123. 04:20over both of those companies and uh a
  124. 04:22fun funny story is today of these 11
  125. 04:25employees seven including myself work
  126. 04:28for GitHub and so different paths Um uh
  127. 04:30they all got to Microsoft and then from
  128. 04:32Microsoft into GitHub. You know coming
  129. 04:34from a small company of of our size
  130. 04:37based in Germany in Stutka Germany,
  131. 04:39southwest Germany. Uh moving you know
  132. 04:42halfway around the world I moved with my
  133. 04:44wife and back then two very young kids
  134. 04:46uh to Seattle that alone was a big
  135. 04:49change. And looking back now um 10 years
  136. 04:51later and we moved we moved in in early
  137. 04:542015 it feels all like a blur like
  138. 04:57things moved so fast and sometimes I
  139. 04:59don't even realize how we did all that.
  140. 05:01Our startup was very small and
  141. 05:02bootstrapped um for the whole time. So
  142. 05:04we never took any outside investment uh
  143. 05:06until Microsoft acquired us and at that
  144. 05:09point in time I think it was about 11 or
  145. 05:11or 12 employees. a still very small
  146. 05:13team. all of all of engineers, you know,
  147. 05:16building building the product together
  148. 05:17and in some ways you know my role today
  149. 05:20as GitHub CEO um is very similar uh in
  150. 05:24that you know um my developer skills uh
  151. 05:27my understanding of code my um you know
  152. 05:30empathy for for how software developers
  153. 05:32work helped me both with my you know
  154. 05:35internal team of about a thousand uh
  155. 05:37engineers and with our customer base
  156. 05:39which are also all software developers
  157. 05:41or those that aspire to become software
  158. 05:43developer And so I think you know a lot
  159. 05:45of the you know my passion for software
  160. 05:47development is actually perfect you know
  161. 05:50for me being being the GitHub
  162. 05:55CEO I don't think I've seen anything
  163. 05:58more exciting and and and changing how
  164. 06:00uh we think about software development
  165. 06:02in my in my 30 plus year career as a
  166. 06:05software developer. You know, when I
  167. 06:07started coding in the early 1990s, uh
  168. 06:10there wasn't even the internet or I I
  169. 06:12certainly had no internet access and so
  170. 06:13I had to figure it all out all out all
  171. 06:15by myself with with books, with
  172. 06:17magazines, um going to computer club in
  173. 06:20the community center kind of hoping that
  174. 06:22somebody will be there. Um if we fast
  175. 06:24forward to where we are now is that it's
  176. 06:26so much easier to get into into software
  177. 06:28development. you can, you know, just
  178. 06:30write a prompt into Copilot or or Chat
  179. 06:32GPT or or similar tools and it uh will
  180. 06:35likely write you um um you know, a basic
  181. 06:38web page um or a small uh application, a
  182. 06:41game uh in Python. And so AI makes
  183. 06:45software development uh so much more
  184. 06:46accessible um for anyone who wants to
  185. 06:48learn coding. And on the other side of
  186. 06:50the spectrum, it makes developers so
  187. 06:51much more productive. Most uh developers
  188. 06:54that uh work on a project have way uh
  189. 06:57too much work to do. uh they have long
  190. 06:59backlogs um of their own ideas of
  191. 07:01customer feedback, you know, uh things
  192. 07:03that they're hearing from their managers
  193. 07:04or from the market or what they're
  194. 07:07seeing at competitors. And so almost any
  195. 07:10software project that has a certain, you
  196. 07:12know, age um has way too much work on
  197. 07:14the innovation side, but they also have
  198. 07:16what we call technical debt. Uh you
  199. 07:18know, legacy code, things that have, you
  200. 07:20know, been created over over months or
  201. 07:22years that need a cleanup, that need
  202. 07:24what we call refactoring. And so
  203. 07:26engineers constantly balance those two
  204. 07:28backlogs. So having something that
  205. 07:29brings the effort down and makes them
  206. 07:31you know 10% 20% maybe even 50% more
  207. 07:34productive is is completely changing uh
  208. 07:37how software developers work. the role
  209. 07:39of GitHub you know in this in the first
  210. 07:41uh let's say you know five years of the
  211. 07:43age of AI um given that we started
  212. 07:45working on GitHub copilot in uh June
  213. 07:472020 right after uh GP3 uh uh was was
  214. 07:51first shown to the world is that we want
  215. 07:53to be you know on the forefront um of uh
  216. 07:55AI code generation we want to provide um
  217. 07:58tools to developer to be more productive
  218. 08:01and more happy uh when when writing code
  219. 08:03because the reality is the dream I think
  220. 08:05for most developers that start their
  221. 08:08their journey um uh as a programmer is
  222. 08:11that they have an idea in their head and
  223. 08:13they're trying to find a way how they
  224. 08:15can can get the fastest from that idea
  225. 08:17to an app or or web page or service,
  226. 08:19right? Like the challenge is not that
  227. 08:21developers uh uh don't have enough
  228. 08:23ideas. The challenge is that you take
  229. 08:25that big idea and you have to break it
  230. 08:27down into small building blocks. And as
  231. 08:28you're working on the first block or the
  232. 08:30first module or the first class or uh
  233. 08:33micros service, whatever it is, you're
  234. 08:34you're realizing that this this idea
  235. 08:36that you have is so much more complex to
  236. 08:39implement than than you thought. And so
  237. 08:41from what became a weekend project, it
  238. 08:43becomes a month-long or sometimes
  239. 08:44year-long project. And many, you know,
  240. 08:46apps that, uh, I wrote as a teenager and
  241. 08:49I know, you know, many of my friends and
  242. 08:51and my employees wrote never get
  243. 08:53anywhere because you you ultimately
  244. 08:54realize it's much more complex than you
  245. 08:56thought and it's not worth spending the
  246. 08:57time on it. And in in the world that we
  247. 08:59live today, you can always, you know,
  248. 09:01download an app from the app store or
  249. 09:02find, you know, some online service that
  250. 09:04does the same thing. So, I think AI
  251. 09:06helps us, you know, to realize the dream
  252. 09:08of taking an idea and implementing it
  253. 09:10much faster. And you see um some of the
  254. 09:12early signs of that where very small
  255. 09:14startups sometimes you know five
  256. 09:16developers um and some of them actually
  257. 09:18only one developers uh uh believe they
  258. 09:20can become million uh if not billion
  259. 09:22dollar businesses by leveraging all the
  260. 09:25AI agents uh that are available to them
  261. 09:27and maybe building their own uh to to
  262. 09:29write code uh to write software much
  263. 09:31faster. Now the flip side of that is
  264. 09:33that you know uh I don't think we're
  265. 09:35anywhere close to a world where you can
  266. 09:36just write a single prompt and say build
  267. 09:38GitHub and then an AI agent builds all
  268. 09:41of the features of GitHub um or even
  269. 09:43just the very basic primitives like
  270. 09:45repository storage you know git storage
  271. 09:47and uh issue tracking um because the the
  272. 09:50decisions um that we as developers as
  273. 09:53engineers as product managers have to
  274. 09:55make to build a complex system like
  275. 09:57GitHub you know thousands if not tens of
  276. 09:59thousands decisions um there's the
  277. 10:01simple ones, right? Like which
  278. 10:02programming language, which which open
  279. 10:04source framework, which cloud to use or
  280. 10:06do we even use a cloud, which operating
  281. 10:08system and and so on. But there's the
  282. 10:10much more complex decisions of how you
  283. 10:12architect the system, you know, are you
  284. 10:14building a monolith and or you building
  285. 10:16microservices? And getting to a point
  286. 10:18where agents can make all these
  287. 10:19decisions and write an app that actually
  288. 10:22is a viable business, you know, finds
  289. 10:23product market fit, has a great user
  290. 10:25experience, and and ultimately generates
  291. 10:27both revenue and profit. uh because any
  292. 10:30business at some point has to get to the
  293. 10:31place where they're making profit and
  294. 10:32and return that profit to the to the
  295. 10:34founders or shareholders that I think
  296. 10:36we're we're quite far away and so we
  297. 10:37need uh engineers uh to do engineering
  298. 10:40stuff. They they need to exercise their
  299. 10:42craft and uh apply systems thinking and
  300. 10:45design and and and build really great
  301. 10:47applications.
  302. 10:51I think the unique thing about GitHub is
  303. 10:53its size and both the love uh that
  304. 10:56developers have for you know our brand
  305. 10:58um for our mascot um the octoat or we
  306. 11:01call it internally Mona um and um the
  307. 11:04reputation uh that GitHub has created
  308. 11:07for itself um since the very early days
  309. 11:09since the I remember the early launch of
  310. 11:11GitHub and and meeting um or seeing um
  311. 11:13Chris one of the founders speaking at
  312. 11:16RailsCon in in Las Vegas in in 2009 and
  313. 11:19then signing up for my own account and
  314. 11:20and started using it. I I was excited
  315. 11:23about using GitHub and now we are in in
  316. 11:262025 and there are still many people
  317. 11:27that uh love GitHub but you know what
  318. 11:30what comes with love is also that you're
  319. 11:32not holding back your criticism and that
  320. 11:34we have 150 million users and uh on the
  321. 11:37platform and so there's at least you
  322. 11:40know a million opinions of what are the
  323. 11:42things we should invest on and what's
  324. 11:44working well and and what's not working
  325. 11:45well and what's the one feature that is
  326. 11:47important you know to that set of users
  327. 11:50while it's not important uh to to me and
  328. 11:52and my product leadership team because
  329. 11:54we have you know our own uh strategy and
  330. 11:56decisions to make. So filtering out the
  331. 11:58signal from the noise and I don't mean
  332. 12:00noise in any negative way just so much
  333. 12:03feedback that we're getting. I remember
  334. 12:05when we when we did the acquisition in
  335. 12:072018 and I joined GitHub and afterwards
  336. 12:09um we sent an email to 10 GitHub users
  337. 12:11and say hey we we're looking into a new
  338. 12:12project we would love your feedback. I
  339. 12:14think we got nine responses of exciting
  340. 12:16users saying if you want to I want to
  341. 12:17provide feedback. If you do that in many
  342. 12:20other companies and startups you get one
  343. 12:21response and the one response is kind of
  344. 12:23like well I have like 10 minutes time to
  345. 12:25to give you some feedback. There's just
  346. 12:27so much um uh uh information that's
  347. 12:29that's coming back to us um on on social
  348. 12:32media and our platform in email and
  349. 12:34support tickets and so on. The second
  350. 12:35piece you know that comes to mind is
  351. 12:37that GitHub um for the longest time has
  352. 12:39been a company with a very strong remote
  353. 12:41culture like long before co uh the
  354. 12:43GitHub founders started hiring um
  355. 12:46developers uh sales folks support folks
  356. 12:48all over the world and today I think
  357. 12:50we're one of the largest remote only
  358. 12:53companies where all work you know from
  359. 12:54our homes or from you know hotel in in
  360. 12:58soul um or from from wherever we are and
  361. 13:00so a lot of um our culture is focused
  362. 13:03around GitHub as a platform form which
  363. 13:05obviously through open source encourages
  364. 13:07um uh asynchronous uh collaboration uh
  365. 13:11and then tools like Slack um um and and
  366. 13:14video video calls um uh that we use much
  367. 13:17more heavily internally than we're using
  368. 13:20email like like old school uh companies
  369. 13:22do. And so when I wake up in the morning
  370. 13:24especially here you know in Seoul uh
  371. 13:26which is like lots of time zones away uh
  372. 13:28from from the US uh where about 80% of
  373. 13:31our employee population is is that I
  374. 13:33wake up to 30 40 Slack messages plus you
  375. 13:36know uh uh hundreds of channels uh with
  376. 13:38with conversation and then figuring out
  377. 13:40you know what is actually important for
  378. 13:41me as CEO what what to react to what can
  379. 13:44I snooze uh for a while and what can I
  380. 13:46just ignore that that's the you know a
  381. 13:49big part of my job but it's also so
  382. 13:50exciting because you know, I can be here
  383. 13:53in South Korea um at at this event and
  384. 13:55and still still run the company. And a
  385. 13:58lot of, you know, what we do on a
  386. 13:59day-to-day basis doesn't actually matter
  387. 14:01whether I'm here in Seoul, or whether
  388. 14:02I'm in Berlin or whether in I'm in New
  389. 14:04York or anywhere else in the world. And
  390. 14:06and I think that's for for many hubbers,
  391. 14:08how we call our employees. A really
  392. 14:10strong part of our culture that we are
  393. 14:12remote company. Uh it's not related to
  394. 14:14the pandemic. It's a choice that we made
  395. 14:17of how we want to run the company. you
  396. 14:19know how we select GitHub as an employer
  397. 14:21and ultimately how we believe we can be
  398. 14:26successful. I first of all I strongly
  399. 14:28believe that every kid uh every child
  400. 14:31should learn coding and we should
  401. 14:33actually teach them coding in school in
  402. 14:34the same way that we teach them physics
  403. 14:36and geography and uh you know literacy
  404. 14:38and math and whatnot. Those are all
  405. 14:40fundamental skills. Coding is one such
  406. 14:42skill and it just has taken us too long
  407. 14:44to actually realize that because
  408. 14:46software is everywhere. Um hardware is
  409. 14:48also everywhere. We carry we carry both
  410. 14:50software and hardware with us uh through
  411. 14:51our day. And in fact our day-to-day is
  412. 14:53dominated by software already. You can't
  413. 14:55really uh live your life uh travel um
  414. 14:58wake up in the morning without software
  415. 14:59anymore. And I think as as humans it is
  416. 15:02crucial to not only be in read only mode
  417. 15:04but also be able to create uh things
  418. 15:07ourselves or at least understand how how
  419. 15:09creation is done on these devices. That
  420. 15:11doesn't mean that every you know 18
  421. 15:1219year-old when they leave high school
  422. 15:14become a software developer. In the same
  423. 15:16way that not every uh uh kid that learns
  424. 15:18physics or or chemistry in school
  425. 15:20becomes a physicist, right? Just because
  426. 15:21you learn those fundamental skills
  427. 15:23doesn't mean that you decide for
  428. 15:24yourself that that's the career path to
  429. 15:26take. So that's number one. You you got
  430. 15:27to you got to learn coding. I think
  431. 15:29number number two is you got to use AI
  432. 15:31to do that. And and you know whether
  433. 15:34it's here in Korea or in Germany uh most
  434. 15:36kids um uh and in fact you know most
  435. 15:39most people don't speak fluent English
  436. 15:41um which is the primary language of
  437. 15:42software development. And so it
  438. 15:44democratizes access to technology and
  439. 15:46that's true for many other uh things in
  440. 15:47the world. And so having uh an agent
  441. 15:50available uh that answers you any
  442. 15:52question but also lets you you know
  443. 15:54realize your dream building your dream
  444. 15:56uh is incredibly exciting. And then the
  445. 15:57third thing you know for anyone who is
  446. 15:59already a software developer or wants to
  447. 16:01you know develop their craft is you got
  448. 16:03to keep you got to keep rehearsing. You
  449. 16:04got to keep training. Yeah. You got to
  450. 16:06keep learning. There's you're never done
  451. 16:08with learning. If I look back 30 years
  452. 16:10of what development looked like back
  453. 16:11then and what it looks like now, um I
  454. 16:13would have been, you know, um very
  455. 16:15behind if I hadn't constantly read blog
  456. 16:17posts, um uh literature and and tried
  457. 16:20out things myself. So I think those are
  458. 16:22uh as as crucial as they were in the '
  459. 16:2490s, they are still crucial in 2025. You
  460. 16:26just have so much more access to
  461. 16:27information uh to become, you know, top
  462. 16:30of the
  463. 16:32field. The obvious answer is that I most
  464. 16:35enjoy using GitHub Copi, right? like
  465. 16:37that's our product. Uh it's our baby. Uh
  466. 16:40we're working on this day in day out. I
  467. 16:42often, you know, uh see features long
  468. 16:44before the world sees that. And and the
  469. 16:46flip side is also true that I often
  470. 16:48don't actually know what's shipped
  471. 16:50versus what is, you know, in in preview
  472. 16:51or in in internal ships. And so I'm
  473. 16:54daily excited about what we're building
  474. 16:56there and I'm using a lot of that myself
  475. 16:57as I am at heart a developer. And
  476. 17:00sometimes it's very simple, you know,
  477. 17:01asking it to to write me a quick script
  478. 17:04that that downloads, you know, ids of
  479. 17:05all our repositories um from our API.
  480. 17:08And in the past, I would have gone, you
  481. 17:10know, to our API documentation and
  482. 17:11figure that out all myself and probably
  483. 17:13would have taken me like half an hour to
  484. 17:15to get to, you know, a shell script that
  485. 17:18does that. Today, I just ask Copil and
  486. 17:20it writes me the script and it works
  487. 17:22within within minutes. Um, and I think
  488. 17:24actually that's the one of the true
  489. 17:25superpowers of AI. Whether that's
  490. 17:27learning to code or exploring the world,
  491. 17:29you have um an assistant available to
  492. 17:32you that has infinite patience. You
  493. 17:33know, it doesn't judge you. Um, it will,
  494. 17:35you know, chat GPT or co-pilot never
  495. 17:37tells you what what a stupid question
  496. 17:39is. It always gives you an answer and it
  497. 17:40even accepts when you tell it it's wrong
  498. 17:42or it needs to, you know, explore the
  499. 17:44topic a little bit further. You know,
  500. 17:46you have seen uh the prompt examples
  501. 17:48where by telling it to outline its
  502. 17:50thought process, it actually gets to a
  503. 17:52better answer. I love using it for my
  504. 17:54blog posts and PowerPoint presentation
  505. 17:56to just uh generate some images and and
  506. 17:59play with that. Um I'm I'm really bad at
  507. 18:02using Photoshop and and and drawing
  508. 18:04myself, but I'm I'm really creative and
  509. 18:06I can write prompts and and figure out
  510. 18:08how to rewrite the prompt to make the
  511. 18:10image look more than uh more more
  512. 18:12closely to what I had in my head.
  513. 18:14There's, you know, tools that we're
  514. 18:15using like teams copilot to summarize
  515. 18:17meetings, especially when I I'm on
  516. 18:18business trips like that. Um, where I
  517. 18:20miss a lot of the meetings that happen
  518. 18:22uh in on the west coast and west coast
  519. 18:24time zone. Just getting a summary,
  520. 18:26figuring out other action items for me.
  521. 18:28Uh, same for summarizing emails. Um, uh,
  522. 18:30you know, using something like reclaim
  523. 18:32AI to to manage my calendar. Those
  524. 18:34things are making me more productive.
  525. 18:35And I think the really exciting thing is
  526. 18:37that there's always a new tool to try
  527. 18:39out and see is that how far along that
  528. 18:42journey is AI and how much more do we
  529. 18:44still have to do as an industry to
  530. 18:46actually get to that to that dream of of
  531. 18:48having an orchestra of agent uh that
  532. 18:50that we're controlling during our
  533. 18:51personal and in our professional lives.
"Don’t Learn to Code" Is WRONG | GitHub CEO Thomas Dohmke — Transcriptly