183 - Struggling To Find Data Jobs? Try This Free Tool I Built

Speaker: 00:00:00

Hey, podcast listeners, Avery Smith here.

2

: 00:00:02

I've been wanting to tell you guys about

this for a while, but like you've probably

3

: 00:00:05

noticed that I've started to make a lot

of my podcast episodes more video heavy,

4

: 00:00:11

and I wish that I could put the videos

on podcast warm, but that's just not how

5

: 00:00:16

it's really set up to work right now.

6

: 00:00:18

Um, so one thing I was thinking

about is how can I make these

7

: 00:00:21

video more video-centric episodes?

8

: 00:00:23

Better for you guys and the listeners.

9

: 00:00:25

Right.

10

: 00:00:26

First off, this one that you're gonna

about to listen to, I don't think

11

: 00:00:28

it's super important to have video.

12

: 00:00:31

Um, I do talk about like a

little bit of a workflow.

13

: 00:00:34

Um, but actually if you go to the

description and you click on the

14

: 00:00:38

link, that will take you to the

workflow that will actually show

15

: 00:00:41

you a picture of the workflow.

16

: 00:00:42

So you could actually have

that open if you wanted to.

17

: 00:00:44

Um, otherwise I like try to explain it

so you don't need the visual reference.

18

: 00:00:48

Um, but I've been thinking about

like these more video centric

19

: 00:00:52

episodes about how can I make it

more special and interesting to

20

: 00:00:55

you guys as podcast listeners.

21

: 00:00:58

And one of the ideas I had, I had was

like, kind of doing this more free

22

: 00:01:01

flowing casual intro to the episodes.

23

: 00:01:05

That you're listening to right

now, that kinda gives you a little

24

: 00:01:07

bit more, uh, behind the scenes.

25

: 00:01:10

And maybe is on a lot of time when you're

making a YouTube video, you have to keep

26

: 00:01:13

it pretty short and brief because people's

attention spans are short, including mine.

27

: 00:01:18

Uh, but when you're doing a podcast, you

can like listen a little bit longer, um,

28

: 00:01:22

and get a little bit more of the details.

29

: 00:01:23

So I was like, maybe I can kind of

do a little preamble to the, uh.

30

: 00:01:27

The actual episode and give

you guys a little bit more

31

: 00:01:30

of an audio only experience.

32

: 00:01:32

So, um, let me know in the Spotify

comments if you like this or not, or

33

: 00:01:37

you can send me an email too and tell

me your thoughts on the podcast, um,

34

: 00:01:41

and maybe how we could do better with

these more video centric episodes.

35

: 00:01:45

You can send me an email, my email's

avery@datacareerjumpstart.com

36

: 00:01:49

and you can just put in the

headline like podcast feedback.

37

: 00:01:52

I'd love to hear from you because it's

something I'm actually thinking about

38

: 00:01:54

is how do we make the audio podcast.

39

: 00:01:57

Continue to be the number one

data podcast on Spotify, which you

40

: 00:02:00

guys, thanks to you guys we are.

41

: 00:02:02

Which is super cool.

42

: 00:02:03

Um, so this episode it's about a

tool I made using a no code slash low

43

: 00:02:08

code, uh, data analytics platform.

44

: 00:02:11

And, um, I explain how I

built it, what the tool is.

45

: 00:02:14

It, it basically, you upload your

resume and it matches you with jobs.

46

: 00:02:17

And I explain kind of why I think no code

and low code tools are, are important.

47

: 00:02:23

But I just wanna tell you guys, I do think

that these tools are really important.

48

: 00:02:26

I think they are going to be

used more in the future and they

49

: 00:02:30

actually make data analysis easier.

50

: 00:02:32

So if you're brand new to data analytics

these tools are, are pretty cool.

51

: 00:02:35

Um, now they aren't used like all the time

in industry, but they're becoming more

52

: 00:02:40

popular and I think they'll be, continue

to become more popular down the road.

53

: 00:02:45

Um, if you're unfamiliar with them,

just think instead of like writing

54

: 00:02:47

code, like SQL or Python or doing

a spreadsheet like Excel or like a

55

: 00:02:53

dashboard like Power BI or Tableau.

56

: 00:02:56

You have like a blank canvas.

57

: 00:02:58

And on that canvas, you, you can

kind of think of it almost like

58

: 00:03:02

a factory, like a data factory.

59

: 00:03:05

Um, where, you know, data starts on

like the left hand side, and then

60

: 00:03:09

your analysis is gonna be on the, the

output of the analysis is gonna be

61

: 00:03:12

on the right hand side and in between

there's like all these like magic

62

: 00:03:17

machines that will do certain things.

63

: 00:03:19

So like, you'll have a data

cleaning machine, right?

64

: 00:03:21

You put the data through the left

hand side through the data cleaning

65

: 00:03:24

machine, and out comes clean data.

66

: 00:03:26

And then you could have like a, a

bar chart machine where you feed

67

: 00:03:29

in the data and out the right

hand side comes a bar chart.

68

: 00:03:32

Um, and you could have all

these different little machines.

69

: 00:03:34

These are called nodes in the tool

I was using, which is called nine.

70

: 00:03:38

Um, these nodes basically perform one

little bite sized data operation for you.

71

: 00:03:43

So it could be, you know,

cleaning all the dates.

72

: 00:03:46

It could be cleaning you

know, the null values.

73

: 00:03:50

It could be doing regression.

74

: 00:03:52

It could be, you know, any sort of

data thing you could be doing can

75

: 00:03:55

be done in this no-code platform.

76

: 00:03:57

So, I think it's really cool.

77

: 00:03:59

Nym this, this pon, this

episode is sponsored by Nime.

78

: 00:04:02

They actually reached out to me and

I was super excited when they reached

79

: 00:04:05

out to me, and I, I talk about this in

the episode, but Trevor, uh, Maxwell,

80

: 00:04:09

who you may have listened to his

interview where he went from no college

81

: 00:04:12

degree printer technician to landing

a totally remote data analyst job.

82

: 00:04:17

He actually uses Nime a decent

amount at work, which is super cool.

83

: 00:04:20

And he's, he's told me how,

how Nims he really enjoys Nime.

84

: 00:04:24

And I was like, yeah, I think

low-code, no-code stuff is cool.

85

: 00:04:27

We don't talk about any of that

inside of the bootcamp, inside

86

: 00:04:29

of my accelerator program.

87

: 00:04:31

Trevor, why don't you make a, um, and I,

I've hired Trevor now, if you didn't know.

88

: 00:04:35

He's one of my coaches inside

of my accelerator program.

89

: 00:04:38

Um, helps manage the students,

answers, questions, does some office

90

: 00:04:41

hours, gives people like a little

bit of an insight of what it's

91

: 00:04:44

like to actually be a data analyst.

92

: 00:04:46

You know, working right now.

93

: 00:04:48

And I had him create, uh, like a

half hour intro to Nime, uh, to our

94

: 00:04:54

accelerator students, so that way if they

see it on job applications or if they

95

: 00:04:56

see low-code, other no-code, low-code

platforms, they would know about it.

96

: 00:05:00

So anyways, uh, big fan of Nime.

97

: 00:05:02

I think it was really cool to have

them reach out and I was super happy to

98

: 00:05:06

tell more people about the tool because

one, I think it's a cool tool and, uh,

99

: 00:05:10

two, I think everyone should be aware

of how to do these types of low code.

100

: 00:05:14

No-code analysis and what

the pros and the cons are.

101

: 00:05:17

I talk a little about the pros.

102

: 00:05:19

In the episode that we'll get

to here just in one second.

103

: 00:05:21

I don't really talk too much of the cons.

104

: 00:05:24

The, the trade off that a lot of people

will tell you is low-code, no-code tools

105

: 00:05:29

make it easier to do data analysis.

106

: 00:05:31

Like it's just easier to set up.

107

: 00:05:33

It's actually easier to

do all the operations up.

108

: 00:05:35

You know what you're trying to do.

109

: 00:05:37

Um, but sometimes they come

at a sacrifice of like speed.

110

: 00:05:42

So I know there are some data

engineers who would argue that like

111

: 00:05:45

low code tools are inefficient and

they really slowed down things from

112

: 00:05:49

a speed and memory perspective.

113

: 00:05:52

I haven't really dealt with these

tools at scale, so I can't really

114

: 00:05:56

speak to how big of a problem that is.

115

: 00:05:59

So if you're currently employed as a

data professional or down the road when

116

: 00:06:02

you, you know, land your first data

job and they're talking about this,

117

: 00:06:05

that's, that's basically a trade off.

118

: 00:06:06

You can, you can know is there, there,

there could be a trade off between how.

119

: 00:06:12

Efficient the tools are

versus how easy it is.

120

: 00:06:14

It's easier to set up, but it

might not be as like efficient

121

: 00:06:17

from a data storage perspective.

122

: 00:06:18

Now, I can't speak specifically

on Nime and like I've never worked

123

: 00:06:21

with, with an a low-code tool

in my personal data per career.

124

: 00:06:26

I'm pretty sure maybe I have a little bit.

125

: 00:06:29

We have a little bit.

126

: 00:06:30

But speed and efficiency

wasn't a big issue for us.

127

: 00:06:33

So if you're in a place where speed

and efficiency are really, really,

128

: 00:06:36

really, really, really important,

so like think about like automatic

129

: 00:06:39

trading on the stock market, right?

130

: 00:06:40

Even one second late, you can be

losing on, you know, a lot of money.

131

: 00:06:44

So those situations, you know, you might

want to lose some of the infrastructure,

132

: 00:06:48

the ease of like a low-code, no-code tool.

133

: 00:06:51

Um, but if you're doing what I was kind

of doing like at ExxonMobil where the

134

: 00:06:55

stuff you're predicting is months out.

135

: 00:06:57

If it takes five extra minutes

to run, it's not a big deal.

136

: 00:07:00

So anyways, that's a little bit of

some background on this episode.

137

: 00:07:04

Tell me in the Spotify comments if

you kinda liked this preamble, uh, to

138

: 00:07:08

the episode that's about to happen.

139

: 00:07:09

And if you want more of it, um, once

again, when I talk about the workflow

140

: 00:07:14

in the upcoming episode, go to the

show notes down below and uh, you can

141

: 00:07:18

see a picture in the download nine.

142

: 00:07:20

Uh, workflow Link and that

will show you the picture if

143

: 00:07:22

you want to have the visual.

144

: 00:07:24

Otherwise, I think I did an okay job

of giving you the visual without,

145

: 00:07:27

without actually showing you the visual.

146

: 00:07:29

Okay.

147

: 00:07:29

Enough of my preamble.

148

: 00:07:30

Let me know if you enjoyed this

and let's get into the episode.

149

: 00:07:32

Okay.

150

: 00:07:33

Speaker: Job hunting sucks right now.

151

: 00:07:35

So in an effort to help you out,

I built a free tool that allows

152

: 00:07:39

you to find specifically tailored

data jobs that are great fits for

153

: 00:07:43

you based only on your resume.

154

: 00:07:46

Seriously, just upload your resume

and boom, you got tailored job

155

: 00:07:50

listings ready for you to apply.

156

: 00:07:52

And I built this entire tool

using a data platform called nym.

157

: 00:07:55

Which allows you to do data analytics

and data engineering with a visual

158

: 00:07:59

interface, which makes the whole process

kind of feel like building data Legos.

159

: 00:08:03

And they're actually the sponsor for

this episode, but more on them in a bit.

160

: 00:08:07

So to test this tool out and to see

if it was any good, I asked one of

161

: 00:08:10

my bootcamp students, you it, if

we could try it with her resume.

162

: 00:08:13

So here's what we do, you use

the link in the show notes

163

: 00:08:16

down below and you're going to.

164

: 00:08:17

Upload your resume.

165

: 00:08:18

So we'll go ahead and upload UITs

resume, then go for the job title

166

: 00:08:22

you want in this case, data analyst.

167

: 00:08:25

Next you're going to say your location

UITs is Charlotte, North Carolina.

168

: 00:08:30

And finally, you're going to give a brief

summary of what your search goals are.

169

: 00:08:33

So for her, it's going to land first.

170

: 00:08:35

Data analyst job, you'll then press

next right here and wait a few

171

: 00:08:40

seconds for it to finish the results.

172

: 00:08:42

And bam, you get your own

personalized dashboard here.

173

: 00:08:45

So on the left hand side, this is where

you have the top 10 job matches for you.

174

: 00:08:50

Uh, and a little bit of

personalized tips right here.

175

: 00:08:53

Then you have the job map here that lets

you see where the different jobs are

176

: 00:08:56

and some basic information about them.

177

: 00:08:58

So if you live in a certain

area, you can check it out there.

178

: 00:09:00

And then on the right hand side.

179

: 00:09:02

This is basically the top 10

jobs that it has for you with the

180

: 00:09:05

descriptions, the location, the

company, as well as the salary.

181

: 00:09:09

And of course, a link to apply.

182

: 00:09:11

You'll need to hit control.

183

: 00:09:12

Click to open that up.

184

: 00:09:14

There's also a fun little histogram

up here that shows you kind of,

185

: 00:09:16

uh, the salary distributions of

what you can kind of expect based

186

: 00:09:19

off of the jobs it found for you.

187

: 00:09:21

Pretty fun way to just find some new

jobs that you could possibly apply for.

188

: 00:09:25

I have a link for you to try this tool

down in the description down below.

189

: 00:09:28

So go try it out and let me know

what you think in the comments.

190

: 00:09:31

I am of course, hoping it'll be

useful for you and help you find

191

: 00:09:34

a few data jobs to apply for.

192

: 00:09:36

But in the meantime, let me show you

how I built it and how you can actually

193

: 00:09:39

build something quite similar, even

if you're not a data expert already.

194

: 00:09:43

To start, I want to tell you about

why I chose to build this tool in

195

: 00:09:47

nine, and you may have never heard of

nine before, and that's totally okay.

196

: 00:09:51

I'll explain what it is and when you

might use it so that way if you ever

197

: 00:09:54

see it on a job description, you don't

have to like panic and be like, oh my

198

: 00:09:57

gosh, I have no idea what this thing

is, can I, I've never even heard of it.

199

: 00:10:00

Right?

200

: 00:10:00

Uh, let's explain what it's, so,

like I mentioned earlier, Nime is

201

: 00:10:03

a data analytics and automation

platform, and it's honestly becoming

202

: 00:10:07

a more popular choice of doing data

analytics and data engineering because

203

: 00:10:11

one, it's free and open source and

we like free, we like open source.

204

: 00:10:15

Who doesn't like that?

205

: 00:10:15

Right?

206

: 00:10:16

And number two, it makes data analytics

and data engineering honestly.

207

: 00:10:19

Fun and easy using nine feels

like playing with data Legos.

208

: 00:10:24

Instead of writing really long coding

scripts, you're actually pushing data

209

: 00:10:27

through a visual workflow with data

building blocks like data Legos, and

210

: 00:10:32

it's a lot easier to do complex data

manipulation because you're doing it with

211

: 00:10:36

like little building blocks and Legos

instead of writing lots of code, which

212

: 00:10:39

of course that makes it more fun and it

makes it a lot easier to do, especially

213

: 00:10:43

if you're not even all that technical.

214

: 00:10:44

You can do this stuff

and it's not too bad.

215

: 00:10:46

Another reason I chose nine is it has

this awesome feature that makes it super,

216

: 00:10:50

super, super easy to share data, web apps

like this resume one that I showed you

217

: 00:10:55

earlier with just a couple clicks, and

that is insanely valuable, uh, because

218

: 00:11:00

that's actually usually pretty hard to do.

219

: 00:11:02

Let me explain why with a quick story.

220

: 00:11:04

So when I was data scientist

at ExxonMobil, I was doing

221

: 00:11:07

cool data science things.

222

: 00:11:08

I was building cool data science projects,

and I like to think that these projects

223

: 00:11:12

were technically difficult, right?

224

: 00:11:13

We were doing cool things

where we were like.

225

: 00:11:15

Predicting gasoline consumption

and trying to figure out what

226

: 00:11:18

oils to buy around the world.

227

: 00:11:19

And we were doing most of these projects

in Python, and it was like hundreds

228

: 00:11:23

and hundreds of lines of Python code.

229

: 00:11:26

We were using all sorts of different

Python libraries, some that you've heard

230

: 00:11:29

of, some you've never heard of before.

231

: 00:11:30

And we were using some decently

complicated machine learning algorithms.

232

: 00:11:34

And so it wasn't easy.

233

: 00:11:35

It was definitely like kind of hard.

234

: 00:11:37

But despite all that.

235

: 00:11:38

The hardest part might have

actually been when we actually

236

: 00:11:41

finished writing all of the code.

237

: 00:11:43

We're like, Hey, we built this really

cool tool, but how do we actually give

238

: 00:11:46

it to end business user who doesn't know

how to use Python, doesn't have Python

239

: 00:11:50

installed in their machine, doesn't

know anything about data, like how do

240

: 00:11:53

we give it to a layman and be like, Hey,

we built this cool tool, now use it.

241

: 00:11:57

Uh, and that was actually a

really hard question to solve.

242

: 00:12:00

You could send them the script and

install Python on their computer, but

243

: 00:12:03

that seems like overly complicated.

244

: 00:12:05

And also, how do they actually run it?

245

: 00:12:06

Do they just click run in like an id?

246

: 00:12:08

It's, it's complicated.

247

: 00:12:09

Do we just give it to a notebook?

248

: 00:12:10

I don't know.

249

: 00:12:10

That doesn't seem like we're

actually solving the problem.

250

: 00:12:12

So, um, this is something that we

actually solved at Exxon most of the

251

: 00:12:16

time by building pretty intensive.

252

: 00:12:18

Web apps, but that required us to

write more Python codes, so we'd solve

253

: 00:12:22

the problem with more Python code.

254

: 00:12:24

And that took a ton of time and a ton

of resources because not only is it a

255

: 00:12:28

lot of coding, but it's also a ton of

data engineering to give us like the

256

: 00:12:31

access and like the CPUs and, uh, I don't

know, hosting it and all that stuff.

257

: 00:12:35

It's, it's kinda over my pay grade, but it

was annoying and it would take forever and

258

: 00:12:39

it would require a lot of people's help.

259

: 00:12:40

Like I couldn't just do it by myself.

260

: 00:12:42

At Exon.

261

: 00:12:43

So with this tool with ny, it has like

this data app feature built right in,

262

: 00:12:47

which makes it super easy to just hand

off to end business users like yourselves.

263

: 00:12:51

So it's kinda like Tableau public.

264

: 00:12:52

If you've ever used Tableau public

before, where you can just create

265

: 00:12:55

a dashboard and share a link right

away, nys the exact same way where

266

: 00:12:57

you can create these dashboards

or web apps and you have a link.

267

: 00:13:00

You just copy that link and you can

literally send it to, uh, anyone you

268

: 00:13:03

want to, and they can use your cool tool.

269

: 00:13:05

So if you or your company are doing

complex data operations with lots

270

: 00:13:09

of data manipulation, lots of data

visualization, and you wanna do it in

271

: 00:13:13

like a simple, fun manner that makes it

easy to share and for easy for people to

272

: 00:13:16

use, nine might be a really good choice.

273

: 00:13:18

And that's when you'd want

to maybe check it out.

274

: 00:13:21

So let me actually explain how you'd

use these data legos and how they work

275

: 00:13:25

here by showing you how I built this

tool and how you could even rebuild

276

: 00:13:29

it yourself or download the workflow

template and then just edit it and

277

: 00:13:33

make some changes and make it your own.

278

: 00:13:34

So basically what you need to know is this

workflow has three to four main parts,

279

: 00:13:39

the resume part right here in yellow.

280

: 00:13:41

The job part here in Orange.

281

: 00:13:44

And then of course the dashboard part

here in red, which is kind of split

282

: 00:13:48

into two with the AI part here, and then

the assembling the dashboard part here.

283

: 00:13:53

Now each one of these little squares

that you see right here is called

284

: 00:13:57

a node, and data flows in the node

from this direction or some sort

285

: 00:14:01

of data transformation occurs.

286

: 00:14:03

And then out the other side to future

nodes downstream as illustrated kind

287

: 00:14:08

of with these arrows right here.

288

: 00:14:10

Now, let's break down each one

of these sections one by one.

289

: 00:14:13

The first part we have right here is

the resume portion, and this is just

290

: 00:14:17

really a node right here, which basically

creates the landing page that you

291

: 00:14:21

saw earlier with all those different

form questions to upload your resume

292

: 00:14:24

and to tell them about yourselves.

293

: 00:14:26

Now, usually building

webpage like this is.

294

: 00:14:29

Pretty difficult to do, but seriously,

not super hard to do In a platform

295

: 00:14:33

like nine where it's all low code,

drag and drop, no coding required.

296

: 00:14:36

Next we have the job board section over

here, which is in orange, and the first

297

: 00:14:40

thing we do here is we have our jobs API

note, which is actually just almost like

298

: 00:14:44

a folder for a whole little mini work.

299

: 00:14:46

Flow inside of the folder.

300

: 00:14:48

Now you'll see here that we're using a

get request node, which just basically

301

: 00:14:51

lets us call our job board API based

on the information that we gave it

302

: 00:14:55

from that landing page, and you can

see some of the data inside of it.

303

: 00:14:58

It's basically pulling the

titles, so on and so forth.

304

: 00:15:01

Then the remaining nodes are

kind of just to clean and

305

: 00:15:04

organized the job listing data.

306

: 00:15:06

Job listing data obviously is

not super structured a lot of

307

: 00:15:08

the time 'cause it's just words.

308

: 00:15:09

It's not, numbers doesn't

fit like in a table, right?

309

: 00:15:11

So this is just kind of us

organizing and cleaning that data.

310

: 00:15:14

And then the remaining nodes inside this

section right here are essentially just

311

: 00:15:18

doing more data cleanup, helping us create

the data that will feed into our LLM.

312

: 00:15:22

We're also doing some data cleaning

down here with regular expressions where

313

: 00:15:26

we're trying to pull out the salaries.

314

: 00:15:28

Inside of the job descriptions and then

report that to you on the dashboard page.

315

: 00:15:32

Now, all the data here basically

gets fed into this section right

316

: 00:15:35

here, which is our AI section.

317

: 00:15:37

The top nodes up here are where you're

going to input your own L-L-M-A-P-I

318

: 00:15:41

keys, which is basically your password

to use some sort of tool like OpenAI

319

: 00:15:45

inside of this app, and then of course

selecting what model you'd like to use.

320

: 00:15:49

In this case, we are

choosing to use GPT-4 0.1,

321

: 00:15:52

and then lastly, we have this

LLM prompter node down here at

322

: 00:15:56

the bottom, which is essentially.

323

: 00:15:57

Our message to chat GPT with

instructions on what to do.

324

: 00:16:00

Now you'll notice with all these arrows,

it's getting our personal resume info,

325

: 00:16:03

the job listing info, and all of the info

from the previous task passed into that.

326

: 00:16:08

All of our information is then fed

into our final section right here,

327

: 00:16:12

which is the dashboard node, and this

is the final page of the web app.

328

: 00:16:15

This is where we have all the settings

to create the actual dashboard.

329

: 00:16:19

Page.

330

: 00:16:19

So to recap, in this workflow, we have

a form that first collects all of the

331

: 00:16:23

user data, a data cleanup section that

extracts information from the resume,

332

: 00:16:28

cleans it up, and then calls a job

board API, an AI section that creates

333

: 00:16:32

custom tips for us based off of all

the previous information collected.

334

: 00:16:35

And then lastly, our user-friendly

dashboards to display all

335

: 00:16:39

that information to the user.

336

: 00:16:40

And this is all built using nym.

337

: 00:16:43

Data building Block Legos.

338

: 00:16:44

So super fun to build

and super easy to do.

339

: 00:16:47

So I hope you now see how it's

possible to do serious data analysis

340

: 00:16:51

workflows using this type of platform.

341

: 00:16:53

And I know a lot of companies are

using tools like Nine to simplify

342

: 00:16:56

their lives, make it easy, and

create cool things like this.

343

: 00:16:59

In fact, one of the alumnis.

344

: 00:17:01

From my bootcamp, Trevor Maxwell,

use his name a decent amount at work.

345

: 00:17:06

You can hear Trevor's full story of going

from printer technician to data analyst

346

: 00:17:10

by clicking on the YouTube card here.

347

: 00:17:12

Or I'll have a link to it

down below in the show note.

348

: 00:17:14

And by the way, if you wanna customize

this workflow and make it your very

349

: 00:17:17

own or make it better, 'cause I

definitely think it could be better.

350

: 00:17:20

You totally can.

351

: 00:17:21

So step one, you just wanna

download this example workflow.

352

: 00:17:24

We'll have a link to it in

the description down below.

353

: 00:17:26

Next, you'd want to download nine

and it's free to get started.

354

: 00:17:29

We'll also have a link to that

down below in the description.

355

: 00:17:31

And then lastly, you'd want to open

that downloaded workflow in nine.

356

: 00:17:35

From here, you can edit any of the nodes

or add more nodes or take away things.

357

: 00:17:39

You can literally do almost anything.

358

: 00:17:41

Just remember, you'll need to enter

your own open AI API and job board API.

359

: 00:17:46

And this is probably the

trickiest part since these are

360

: 00:17:48

third party external tools.

361

: 00:17:50

So that would be a little bit tricky.

362

: 00:17:51

But other than that, you could

literally do so many things.

363

: 00:17:54

You could increase the

number of jobs returned.

364

: 00:17:56

You could add some analysis about

what skills are mentioned the most.

365

: 00:17:59

Maybe you could even create some

sort of a scoring system, like

366

: 00:18:02

there's so many different things

you could be doing with this guys.

367

: 00:18:05

So go ahead, try it out.

368

: 00:18:07

Overall, I hope you now one,

have a fun tool to test, to try

369

: 00:18:11

to find different data jobs.

370

: 00:18:12

Two, you understand how you

can actually analyze real data.

371

: 00:18:15

With low code building blocks.

372

: 00:18:17

Legos, right?

373

: 00:18:18

It's a lot of fun.

374

: 00:18:18

And three, I hope you know what

nine is, so you won't be scared

375

: 00:18:21

when you see it on a job description

and you can be like, oh yeah, Avery

376

: 00:18:24

talked about that in an episode.

377

: 00:18:26

I'm hoping all of this is going

to help you on your data journey,

378

: 00:18:29

and if you want more help on

your data journey, hit subscribe.

379

: 00:18:31

I've got a lot more useful tips to share.