YouTube2Text

Should You Still Become a Data Analyst in 2027? | Microsoft VP — Transcript

by Sundas Khalid · 4,350 words · 659 segments · language en · Watch on YouTube

Full transcript

  1. 0:00I don't think you need to write a single
  2. 0:01line of SQL anymore.
  3. 0:03>> This is Priya Sathi, vice president of
  4. 0:05SQL at Microsoft. With nearly 30 years
  5. 0:08of experience in data and analytics.
  6. 0:09>> You really need to know SQL so you're
  7. 0:12able to unlock the data, bring AI to the
  8. 0:15data, and now you're able to get
  9. 0:17insights. Do you need to write SQL?
  10. 0:20Well, probably not, right? But do you
  11. 0:22need to learn SQL? Yes.
  12. 0:23>> But guess what? Knowing SQL is only the
  13. 0:25beginning. The next step is
  14. 0:27understanding how to use AI agents to do
  15. 0:29the work that once [music] took analysts
  16. 0:31hours.
  17. 0:32>> I tell everybody that is about using
  18. 0:35agents and knowing how to wield agents
  19. 0:36[music] to give you the superpower.
  20. 0:38>> When I write SQL, it would take me like
  21. 0:40at least like two to three hours. The
  22. 0:42part that shocked me is it was able to
  23. 0:44figure out where the data lives, where
  24. 0:46the tables are. [music] It was able to
  25. 0:48like figure out the business context and
  26. 0:49logic and was able to write very super
  27. 0:53awesome SQL. But there is a bigger
  28. 0:55challenge. When AI agents start working
  29. 0:58with critical business data, how do you
  30. 1:00make sure that their answers are
  31. 1:02actually correct? Somebody who's
  32. 1:04starting their career in analytics,
  33. 1:06should they learn SQL, Python, or
  34. 1:08working in Excel first?
  35. 1:09>> I really think you got to learn that's
  36. 1:11the foundation of everything.
  37. 1:15>> Hi everyone, welcome back to another
  38. 1:16episode of Out of Office with Sunder. In
  39. 1:18today's episode, we have Priya Sathi,
  40. 1:21who is the vice president of SQL at
  41. 1:24Microsoft. So when I learned about that,
  42. 1:26I was really excited to talk to Priya
  43. 1:28because I know you all want to learn
  44. 1:30about SQL, data analytics, agentic
  45. 1:32analytics, and so on. And we're going to
  46. 1:34cover all of that.
  47. 1:35>> Yes.
  48. 1:36>> Priya, thank you so much for being on my
  49. 1:37channel and I'm really excited to talk
  50. 1:39to you today. First of all, I want to
  51. 1:41say you have spent about 30 years in the
  52. 1:43industry data analytics space.
  53. 1:46From your POV, how has industry changed,
  54. 1:49whether that is like data analyst job,
  55. 1:51data scientist job, data engineer job?
  56. 1:53Is this still the sexiest job out there?
  57. 1:56>> Oh my gosh, I am so glad you're asking
  58. 1:58this and I've such an amazing
  59. 2:00transformation. 30 years is a long time.
  60. 2:04>> Right.
  61. 2:04>> And I started my journey. I started my
  62. 2:07journey actually with databases.
  63. 2:09>> Databases.
  64. 2:10>> And uh you know early forms of you know
  65. 2:14building applications on databases to I
  66. 2:17you know
  67. 2:18I went to the journey of building data
  68. 2:20warehouses and got into analytics.
  69. 2:22>> Wow.
  70. 2:22>> And it was all very basic reporting you
  71. 2:25know just trying to get people to you
  72. 2:27know build Excel based reports.
  73. 2:29[laughter]
  74. 2:29You know to a point in time where I am
  75. 2:32so excited right now to be in this era
  76. 2:35of AI because
  77. 2:37it is just the biggest transformation
  78. 2:40I've seen in my career journey of 30
  79. 2:42years with AI not just improving
  80. 2:45productivity but I love AI for the good
  81. 2:49and where we can take AI into building
  82. 2:52solutions that are going to solve real
  83. 2:55problems that would be impossible to
  84. 2:57solve otherwise.
  85. 2:59Um so I'm super excited about where this
  86. 3:02transformation is going towards.
  87. 3:04>> Yeah. Um so if you were to like put if
  88. 3:07we were to like ground this in actual
  89. 3:08skill set 30 years ago versus now, what
  90. 3:11do you think has changed? Maybe we can
  91. 3:13take data analyst as example. Like how
  92. 3:15has those skill set evolved over the
  93. 3:17years?
  94. 3:18>> I think you know coming with the SQL
  95. 3:20background, right? A data analyst even
  96. 3:22back in the late 90s and early 2000s
  97. 3:26would probably be writing SQL code to
  98. 3:28build those reports. You know you're
  99. 3:30probably writing a lot of SQL or you're
  100. 3:32trying to look at data, trying to make
  101. 3:35sense out of this data that's coming out
  102. 3:37of the operational systems and you're
  103. 3:38trying to figure out you know what is
  104. 3:40the next thing I need to do. I think the
  105. 3:43the the things that have stayed intact
  106. 3:47and will continue to compound and stay
  107. 3:48intact is the thing around okay I need
  108. 3:51to now think would this insights I'm
  109. 3:54getting, right? I need to now go figure
  110. 3:58out what to do next, right? I I those
  111. 4:01core SQL critical thinking skills,
  112. 4:03getting you know, the insights and then
  113. 4:06thinking about how those insights
  114. 4:08transform my business still apply, you
  115. 4:11know. It's just in the agentic world, I
  116. 4:13don't think you need to write a single
  117. 4:14line of SQL anymore.
  118. 4:16>> But do you still need to learn SQL?
  119. 4:18>> I think I think really yes, because at
  120. 4:22the end of the day, you know, let's talk
  121. 4:24about why SQL is important, right? SQL
  122. 4:27Server has been around my gosh, 35 over
  123. 4:3035 years, right? And customers have
  124. 4:34built mission critical applications.
  125. 4:37SQL's been the system of record that is
  126. 4:39running ERP, CRM, order management,
  127. 4:43hospital systems. So any industry around
  128. 4:46the world, right?
  129. 4:47>> Wherever the data is. The data is, yeah.
  130. 4:49>> it's living in SQL, right? So if we
  131. 4:53think about bringing that into agentic
  132. 4:56transformation,
  133. 4:58you really need to know SQL so you're
  134. 5:01able to unlock the data, bring AI to the
  135. 5:04data, and now you're able to get
  136. 5:06insights.
  137. 5:07Do I Do you need to write SQL, T-SQL?
  138. 5:11Well, probably not, right? But do you
  139. 5:13need to learn SQL? Yes, because the
  140. 5:15database is housing the very critical
  141. 5:18data.
  142. 5:18>> Right. Right. So if you're trying to
  143. 5:20figure out if you should learn SQL or
  144. 5:22not learn SQL, the answer is yes. So
  145. 5:25otherwise you're not going to know
  146. 5:26whatever AI is spitting out is correct
  147. 5:28or not.
  148. 5:28>> Yeah, exactly.
  149. 5:29>> But you also touched on a very important
  150. 5:31uh point like okay, knowing the SQL is
  151. 5:34good enough. There has to be another
  152. 5:35step now that there is AI present. What
  153. 5:38exactly is that step?
  154. 5:39>> The The I think the next step is really
  155. 5:42understanding how to use agents. And I
  156. 5:44tell
  157. 5:46I tell everybody that is about using
  158. 5:48agents and knowing how to wield agents
  159. 5:50to give you the superpower. It's like
  160. 5:52agents make you
  161. 5:54the superhuman, right? And that is
  162. 5:57really where, you know, how to use
  163. 5:59agents on top of the data to give you
  164. 6:02the results that you need, making sure
  165. 6:04it's not garbage in, garbage out,
  166. 6:05because that's the the foundation, you
  167. 6:07know? And you definitely need to know,
  168. 6:11you know, how to ask the right
  169. 6:13questions, for example. That is critical
  170. 6:16thinking that doesn't change, right?
  171. 6:18Understanding the metadata, asking the
  172. 6:21critical questions, those things don't
  173. 6:23change.
  174. 6:23>> Yeah. So I think which um if you were
  175. 6:26like putting it in on the ground, there
  176. 6:27is like a data
  177. 6:29project workflow. You start with data
  178. 6:31access, data cleaning,
  179. 6:32>> Yeah.
  180. 6:33>> uh finding like the business questions,
  181. 6:35building reports and dashboard and
  182. 6:37sharing a storytelling. Where do you
  183. 6:39think like
  184. 6:40agentic analytics come into play between
  185. 6:42like when we're talking about that
  186. 6:43entire workflow?
  187. 6:45>> I think many steps in that workflow can
  188. 6:47be done by agents today, right? It's
  189. 6:50really important for humans to review.
  190. 6:52It's really important for humans to set
  191. 6:54the guardrails, right? At at Microsoft,
  192. 6:56we do make sure that we follow
  193. 6:59responsible AI. It's safe,
  194. 7:02privacy, security, or compliance all are
  195. 7:04taken care of. And in SQL, we took it
  196. 7:08very importantly because we said, you
  197. 7:10know what? Agents are going to access
  198. 7:12the very critical data, sensitive data.
  199. 7:13We have social security numbers in
  200. 7:15databases. We have patient critical
  201. 7:18records in databases, right? That means
  202. 7:20an agent accessing the data
  203. 7:22should absolutely be treated like a
  204. 7:24human accessing the data, right? Data
  205. 7:27use policies, all of this, making sure
  206. 7:30that agents are treated like humans are
  207. 7:34auditable. And uh at the at the end of
  208. 7:37the day,
  209. 7:39we I really think that, you know,
  210. 7:42a lot of these steps agents can do, but
  211. 7:45then humans are setting those
  212. 7:47guardrails, reviewing it, making sure
  213. 7:49that they can then
  214. 7:51get those insights and then, you know,
  215. 7:54help take the next best action for their
  216. 7:58company, for, you know, their
  217. 8:00enterprises.
  218. 8:01>> Yeah. No, that so basically it's every
  219. 8:04step. Every step is certain steps maybe
  220. 8:06you need to be like having more guard
  221. 8:08rails. And in other steps, maybe you
  222. 8:10need to be more involved in terms of
  223. 8:13like whether that is storytelling, like
  224. 8:14having that business context and domain
  225. 8:16knowledge. Um I was working full-time as
  226. 8:19a data scientist up until like 4 months
  227. 8:20ago. And I have seen like the AI
  228. 8:23evolution. I think the biggest aha
  229. 8:25moment for me where I was like
  230. 8:27really amazed where I typically when I
  231. 8:30write SQL, it would take me like at
  232. 8:32least like 2 to 3 hours like from hand
  233. 8:34from
  234. 8:35like for complex tables because the
  235. 8:36biggest part for my job was figuring out
  236. 8:38where the data lives and then writing
  237. 8:40the SQL, then validating it, and then
  238. 8:42making sure the data is correct.
  239. 8:44Um we had AI and I just put in the AI. I
  240. 8:46said, can you help me get this data set?
  241. 8:50Um the part that shocked me is it was
  242. 8:53able to figure out where the data lives,
  243. 8:54where the tables are. It was able to
  244. 8:57like figure out the business context and
  245. 8:58logic and was able to write very super
  246. 9:02awesome SQL and that was my biggest aha
  247. 9:06moment. I was like, oh my god, this is
  248. 9:08insane. I never thought this was
  249. 9:10possible.
  250. 9:10>> Yeah, and you know what? I
  251. 9:12I think most people that end up writing
  252. 9:13SQL don't think about the performance
  253. 9:16impact and you write a query and you
  254. 9:19forget a predicate, you forget a filter,
  255. 9:21you you're not thinking about like you
  256. 9:23know putting an index on the table.
  257. 9:26But AI can do all of that and make sure
  258. 9:29that you get the most performance query,
  259. 9:31right? And so then you're not sitting
  260. 9:33there saying, why is this thing taking
  261. 9:35so long to execute? You are getting your
  262. 9:38results so fast, so all of that
  263. 9:40performance overhead is autonomously
  264. 9:43done.
  265. 9:43>> Yeah. Yeah. Um on that note, I do want
  266. 9:45to ask you, do you how are you using
  267. 9:48agentic analytics? What do you like
  268. 9:50you have a revenue question, do you go
  269. 9:51to your AI agent and be like
  270. 9:53>> do. I do.
  271. 9:55>> [laughter]
  272. 9:56>> So here on the SQL team, we have some of
  273. 10:00the most amazing telemetry and not just
  274. 10:02revenue, right? Like I yeah, I want to
  275. 10:04understand, you know,
  276. 10:06what are people doing with SQL,
  277. 10:09amazing telemetry. And I want to ask a
  278. 10:11telemetry question, I just go to my
  279. 10:14agent. We have amazing agent skills that
  280. 10:17and I just ask the question of the day
  281. 10:19that's top of mind. I wake up in the
  282. 10:21morning and I want to understand whether
  283. 10:23it's revenue, usage,
  284. 10:26you know, customer satisfaction, you
  285. 10:28know, I want to review surveys. I don't
  286. 10:31do it with the old-fashioned way
  287. 10:33anymore. I have either agent skills or
  288. 10:37agents with their briefings ready for me
  289. 10:40every morning that says, oh, good
  290. 10:42morning, Priya.
  291. 10:44>> Okay, you you touched on like the
  292. 10:45old-fashioned way. What was the
  293. 10:47old-fashioned way?
  294. 10:48>> The old-fashioned way, actually even
  295. 10:49until like a couple of years ago for me
  296. 10:52was to go to that report, click on the
  297. 10:54dashboard, refresh and say, oh, wait for
  298. 10:57it to refresh and then say, what what's
  299. 10:59the results? You know, but I that is,
  300. 11:03you know, I'm not going back the very
  301. 11:04old-fashioned way, you know, but
  302. 11:06>> And I'm guessing like you trust that
  303. 11:08what agents is telling you because it's
  304. 11:10coming from like a pre-built dashboard
  305. 11:12that is already
  306. 11:13like basically vetted out by humans and
  307. 11:15so on. Okay.
  308. 11:16>> That that that is. And, you know, we
  309. 11:19have made sure that the skills have been
  310. 11:21vetted out by the agent that's able to
  311. 11:23go look at that information and, you
  312. 11:26know, we have done the
  313. 11:28pre-testing and, you know, so and that,
  314. 11:31you know, I did that myself, so I'm
  315. 11:33making you know, I know that I'm not
  316. 11:35getting wrong results every day. So.
  317. 11:38>> Yeah. So for people who are people who
  318. 11:40are like students, first of all, I have
  319. 11:42two kids. They are both in college and
  320. 11:45none of their schools are actually
  321. 11:46teaching them AI. So a lot of my
  322. 11:48audience who is watching, their schools
  323. 11:50are probably not teaching them AI as
  324. 11:52well.
  325. 11:53How do you actually learn AI? Like how?
  326. 11:56Where do you start in terms of like
  327. 11:57agentic analytics? Like that's the
  328. 11:59focus, not just like general purpose
  329. 12:01LLM.
  330. 12:02>> I I think there's
  331. 12:05Okay, first of all, nobody taught me AI
  332. 12:06as
  333. 12:07[laughter]
  334. 12:09[gasps] a few years ago when I think it
  335. 12:11was ChatGPT that came out and, you know
  336. 12:13everybody just you know started trying
  337. 12:15it out and it was one of those organic
  338. 12:18things that you learned.
  339. 12:20Thing is it's easy because first of all
  340. 12:24you got to choose your make your agent
  341. 12:26of choice, right? And the rest of it is
  342. 12:31you're really having a conversation like
  343. 12:33you're having a conversation with your
  344. 12:34daughter, right? And you're a parent,
  345. 12:37I'm a parent as well and you know we
  346. 12:39really I think about AI as literally
  347. 12:42that conversation with my daughter
  348. 12:44telling her exactly what I want, setting
  349. 12:47the guard rails. She's a teenager, you
  350. 12:48know? So you know
  351. 12:50>> how old is your daughter?
  352. 12:51>> Oh she's 13.
  353. 12:52>> 13 okay.
  354. 12:53>> Yes. And so giving her instructions,
  355. 12:55setting the guard rails, making sure I
  356. 12:57know what she you know can and cannot
  357. 13:00do. You're right, right? It's important.
  358. 13:03But that's the conversation I'm having
  359. 13:05and I I have to be very specific, you
  360. 13:07know? I don't want to leave room for
  361. 13:08interpretation, right? And the better I
  362. 13:12am at that I get better results.
  363. 13:14>> Yeah. No I love that analogy and if you
  364. 13:17have kids you can definitely like now
  365. 13:19take that and translate that into your
  366. 13:21work. If you don't have kids just think
  367. 13:22of it like your intern or your assistant
  368. 13:24or your little brother or sister or
  369. 13:26cousin or whatever that is. Basically
  370. 13:28giving them instructions to do something
  371. 13:29and just add data to the mix and ask
  372. 13:32them to do like data projects.
  373. 13:34When I say them I mean agentic
  374. 13:36analytics.
  375. 13:37Give them instructions, guard rails and
  376. 13:38what not. So very cool. Thank you for
  377. 13:40sharing that analogy. Now I do want to
  378. 13:42pivot a bit and talk about your career
  379. 13:44journey. I know you come from a small
  380. 13:47town. Could you walk us through like how
  381. 13:50in the world did you come from a small
  382. 13:52town to now being a vice president of
  383. 13:55none other than SQL,
  384. 13:57my favorite language?
  385. 13:59How did you do it?
  386. 14:01>> Okay. So I've been about 30 years in the
  387. 14:04industry and yes my journey begins in a
  388. 14:08small town in India and I didn't imagine
  389. 14:12that this is where I would end up. I
  390. 14:15just, you know, kind of grew up not
  391. 14:18knowing what I wanted to be.
  392. 14:21I think for the longest period of time I
  393. 14:23thought I wanted to be a doctor because,
  394. 14:25you know, that was the thing to do.
  395. 14:27>> Yeah.
  396. 14:27>> Uh and then I accidentally fell into
  397. 14:29computer science because
  398. 14:30>> What is accidentally?
  399. 14:31>> Well, I kind of didn't get into medical
  400. 14:33school.
  401. 14:34>> [laughter]
  402. 14:36>> Okay, that [gasps] makes sense.
  403. 14:37>> And I chosen I went into the next best
  404. 14:39option, right? And it was very, you
  405. 14:42know, back then, you know, talking 30
  406. 14:44plus years ago. So, you know, it was
  407. 14:45like the this,
  408. 14:47you know, thing that was everybody was
  409. 14:49talking about. And, you know, there
  410. 14:52weren't a lot of women in this at all
  411. 14:54back then, you know, engineering. And I
  412. 14:58I loved it. I fell in love with it. And
  413. 15:01I felt like I connected with computers
  414. 15:03and you know, I did computer
  415. 15:04engineering, which is really electronics
  416. 15:07plus computer science. So really I built
  417. 15:09my own motherboards and, you know, I
  418. 15:12learned to write assembly language code
  419. 15:14and, you know, really tell computers
  420. 15:16what I wanted them to do. So down in the
  421. 15:19roots of it.
  422. 15:21But, you know, that is where I began my
  423. 15:25journey and I never took no for an
  424. 15:27answer and I didn't want to do
  425. 15:30everything that was, you know, very
  426. 15:32typically what, you know,
  427. 15:35a woman did, right? Back in the day. And
  428. 15:38I said, you know what? I want to go
  429. 15:41explore, expand because I think I'm, you
  430. 15:44know, I'm worth it. And I hopped on a
  431. 15:48plane and I came to the United States.
  432. 15:51And
  433. 15:52I got a master's degree. And I I think
  434. 15:57my career started from right there in
  435. 16:00the late 90s. And I started with
  436. 16:03databases and, you know, now I am with
  437. 16:06SQL.
  438. 16:06>> Did you know back then that you would be
  439. 16:09vice president at Microsoft?
  440. 16:11>> I didn't, but here is the the thing I
  441. 16:14will say. This is what I say is destiny.
  442. 16:17So when I started my first job out of uh
  443. 16:21grad school,
  444. 16:23I did start working with databases and
  445. 16:26uh you know I didn't I wasn't working
  446. 16:28with Oracle back then. I started working
  447. 16:30with uh sorry with SQL back then, but I
  448. 16:31was working with Oracle and then I
  449. 16:35you know moved into SQL and never in the
  450. 16:37wildest dreams imagination did I ever
  451. 16:41think that someday I would be vice
  452. 16:43president of SQL at Microsoft and you
  453. 16:47know it was I started there and destiny
  454. 16:51kind of took me a full big circle all
  455. 16:54the way around 30 years later.
  456. 16:56>> Wow. Was there a specific point where
  457. 16:58you're like I made it?
  458. 17:00I think I made it.
  459. 17:03>> Uh I think that point was actually a few
  460. 17:06years ago, not even before I got to this
  461. 17:09point as vice president. Honestly,
  462. 17:11a few years ago I did get to a point
  463. 17:15where I just looked back at you know
  464. 17:18just all of these amazing experiences.
  465. 17:21My career I call it an experience,
  466. 17:23right? It's a journey of 30 years and
  467. 17:26every single year, every single chapter
  468. 17:29has been an amazing experience.
  469. 17:31>> Right.
  470. 17:32>> Amazing positive or negative, you know,
  471. 17:34but uh even those negative experiences
  472. 17:37have just made me bigger, stronger,
  473. 17:39better. Um and I've always had this
  474. 17:43I don't take no for an answer and
  475. 17:45everything is possible. That is
  476. 17:47something very deeply rooted in me from
  477. 17:50when I was a little girl. I I was told
  478. 17:53you can't do it Priya and I would say
  479. 17:55well I'm going to go prove I can do it.
  480. 17:57>> [laughter]
  481. 17:58>> Wow.
  482. 17:59Are you guys hearing that? Basically
  483. 18:00don't take no for an answer. You know uh
  484. 18:03this is what I'm noticing cuz I was
  485. 18:04talking to uh Ben Goertzel.
  486. 18:07She's also vice president at product
  487. 18:09marketing Microsoft and uh I asked her
  488. 18:12same question. I said like what exactly
  489. 18:14did it take and she said I did not take
  490. 18:16no for an answer. It was persistence and
  491. 18:18obviously being rock star at your job.
  492. 18:20So, uh I'm amazed that you all share
  493. 18:23similar characteristics and I guess like
  494. 18:25people who are watching, if you're
  495. 18:27trying to make it, those are the
  496. 18:28characteristics that make you
  497. 18:30>> Yeah.
  498. 18:31>> make you far, take you far.
  499. 18:32>> biggest thing that stops anybody, and I
  500. 18:34always tell anybody who asks me this
  501. 18:36question, biggest thing that always
  502. 18:38stops anybody from progress is
  503. 18:40themselves.
  504. 18:41>> Wow.
  505. 18:42Did you have a community like a do you
  506. 18:45have a
  507. 18:46I know like going working having a
  508. 18:48working career in tech with kids um it's
  509. 18:52not easy.
  510. 18:53>> Mm-mm.
  511. 18:53>> So, did you have that support system
  512. 18:55whether that is at home or at work,
  513. 18:57whether those are mentors or sponsors
  514. 18:59like how did that look like?
  515. 19:01>> I have the most amazing husband on the
  516. 19:03planet, I would say.
  517. 19:05He probably meets me more than halfway.
  518. 19:08I do you know I'm here for the week in
  519. 19:11Barcelona, right? He's taking care of
  520. 19:14everything at home,
  521. 19:15drop off, pick up, shuttling around and
  522. 19:18doing all of that that we moms do,
  523. 19:21right? And he's been the biggest
  524. 19:23supporter of my career and I would say
  525. 19:26you know I you know would not be able to
  526. 19:29do this without him. But even going back
  527. 19:32into the early stages of my
  528. 19:34you know even when I was a student, you
  529. 19:36know when I didn't get into med school
  530. 19:38and then I ended up in computer
  531. 19:40engineering, it was my dad that pushed
  532. 19:42me and said okay so what you didn't get
  533. 19:44to into medical school. This is a great
  534. 19:47option and I think you're going to love
  535. 19:49it. So, the big thing is these people
  536. 19:52see something bigger than me, bigger in
  537. 19:55me and that I actually maybe don't see
  538. 19:59sometimes, right? I look at the face in
  539. 20:01the mirror but I don't necessarily
  540. 20:02always recognize that I can do it but
  541. 20:06they seem to know me better than me.
  542. 20:08>> Mm-mm.
  543. 20:08>> And that has always worked really great
  544. 20:10for me. And along my career journey,
  545. 20:13I've had some great mentors both you
  546. 20:16know male and female mentors along the
  547. 20:19way. The biggest thing they always saw
  548. 20:21they caught on to okay this is not the
  549. 20:23girl you can say no to, right? She's
  550. 20:25going to just go figure it out. But then
  551. 20:28the biggest thing they always saw is
  552. 20:29they saw something deeper in me that I
  553. 20:31probably didn't recognize. And they said
  554. 20:34that to me knowing that I wouldn't take
  555. 20:36no for an answer and it pushed me into
  556. 20:40you know, achievements that I didn't
  557. 20:42think I would
  558. 20:43>> Yeah. That's beautiful. Thank you for
  559. 20:46sharing that. You having your partner,
  560. 20:48having your father, having the mentors
  561. 20:51and sponsors throughout your career,
  562. 20:53um and being you, awesome, like rock
  563. 20:56star, um that's what it takes. So find
  564. 20:59that support system, find that community
  565. 21:02and be good at what you do. So that's
  566. 21:04the baseline.
  567. 21:05>> True.
  568. 21:06>> so if there is a girl, small town in
  569. 21:08India, Philippines,
  570. 21:10watching this right now,
  571. 21:12and she has big dreams, what would you
  572. 21:14tell her?
  573. 21:16>> I would say the most important thing is
  574. 21:19to believe in yourself. I I'm not
  575. 21:21kidding you when I say the biggest thing
  576. 21:25that prevents you from making your
  577. 21:26achievements is yourself, your internal
  578. 21:30fear that you think you can't do it. You
  579. 21:34have to believe in yourself and know
  580. 21:36that you can do it. That is what unlocks
  581. 21:39your dreams, right? Follow your dreams,
  582. 21:42stay persistent, and if you fall down
  583. 21:44one time, that's okay. You pick up and
  584. 21:47you move on, right? It is We all fall.
  585. 21:51We are not perfect, right? But you pick
  586. 21:54up and you learn from what made you fall
  587. 21:56and you keep persistent and you keep
  588. 21:59moving on. You need to believe in that.
  589. 22:01And that is the only thing I would say
  590. 22:04if you're, you know, Philippines, India,
  591. 22:06wherever you're at, right?
  592. 22:09That is the most important thing.
  593. 22:10>> I love it. That's giving me goosebumps.
  594. 22:12[laughter]
  595. 22:13Thank you for sharing that. Okay, so
  596. 22:15we're getting towards the end of our
  597. 22:16conversation and every episode or every
  598. 22:19guest that I have, I ask I do like a
  599. 22:21rapid fire round where like I have like
  600. 22:24I'm sure you have seen the shows where
  601. 22:25they do rapid fire. So the idea is that
  602. 22:27I'll ask you and then you have to like
  603. 22:29whatever comes to mind. You have to like
  604. 22:31say it really fast. And I haven't shared
  605. 22:33these questions. I'm sorry [laughter] I
  606. 22:34haven't shared these questions with you.
  607. 22:36But like
  608. 22:36>> Oh, that sounds fun.
  609. 22:37>> Okay. Sounds good. I love that. I love
  610. 22:39the energy.
  611. 22:43What is the biggest piece of advice you
  612. 22:44would give yourself 30 years ago?
  613. 22:46>> Oh, going back in time 30 years ago?
  614. 22:50Uh try anything and everything at least
  615. 22:52once.
  616. 22:53No regrets.
  617. 22:54>> A dashboard in Excel or Power BI?
  618. 22:57>> Power BI.
  619. 22:58[laughter]
  620. 23:01[gasps] I have soft spot. [laughter]
  621. 23:03Soft spot for
  622. 23:04>> spot for Power BI. One thing that you
  623. 23:05would not give AI to do?
  624. 23:08>> My taxes, my personal bank accounts,
  625. 23:11anything to do with money.
  626. 23:14>> Somebody who is starting their career in
  627. 23:15analytics, should they learn SQL,
  628. 23:17Python, or working in Excel first?
  629. 23:20>> I really think you got to learn SQL.
  630. 23:22It's the foundation of everything. SQL
  631. 23:25was there before Excel. SQL was there
  632. 23:27before Python. SQL
  633. 23:29was there, has been there for a very
  634. 23:31long time.
  635. 23:31>> Yeah.
  636. 23:32Awesome. Well, these were my questions.
  637. 23:34See, easy. They were not that difficult.
  638. 23:36Not at all. It was fun. It was fun.
  639. 23:38Okay. Is there any question that you
  640. 23:40wish I had asked you?
  641. 23:42>> No, I think this uh this was a lot of
  642. 23:44fun.
  643. 23:45>> Okay. All right. Well, thank you so
  644. 23:47much, Priya. It was so fun talking to
  645. 23:49you and honestly, I feel so inspired uh
  646. 23:52coming from like a small town to now
  647. 23:54making it big. Please keep sharing what
  648. 23:57you're doing and how you have done it
  649. 23:59cuz I 'm sure there are a lot of people
  650. 24:00watching looking for inspiration. And
  651. 24:02yes, you got to learn SQL first when you
  652. 24:04start cuz SQL is the foundation. And
  653. 24:08with that, thank you so much for being
  654. 24:10here. And hopefully all of you guys
  655. 24:12learned a lot from today's conversation.
  656. 24:14If you did, let us know in comments what
  657. 24:17exactly was your biggest takeaway. And
  658. 24:19with that, I will see you in the next
  659. 24:20video. Have a good one. Bye.

About this transcript

This page contains the full transcript of Should You Still Become a Data Analyst in 2027? | Microsoft VP by Sundas Khalid, generated from the public captions YouTube serves with the video. The transcript has 4,350 words across 659 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.