210: Build a Data Analyst Portfolio in 9 Minutes (Full Tutorial)

Speaker: 00:00:00

This is my brother Graham.

2

: 00:00:01

Hi.

3

: 00:00:01

And Graham wants to

land his first data job.

4

: 00:00:03

Yep.

5

: 00:00:04

But he doesn't have a portfolio

that's gonna convince a hiring

6

: 00:00:06

manager to take a chance on him.

7

: 00:00:08

So we're gonna build him a portfolio from

scratch today to having a full working

8

: 00:00:12

portfolio in less than 20 minutes.

9

: 00:00:14

Sound good?

10

: 00:00:14

Sounds great.

11

: 00:00:15

Let's get into it.

12

: 00:00:16

Okay, the tool we're going to be using

today to build a portfolio from scratch

13

: 00:00:19

is called MyDatafolio, and it's a new tool

that lets you build a really beautiful

14

: 00:00:24

portfolio website pretty dang quickly.

15

: 00:00:26

And actually, full disclosure,

it's actually made by me.

16

: 00:00:29

And it's what I would like

to have in a data portfolio.

17

: 00:00:32

So link in the description

down below to try it out.

18

: 00:00:34

All right.

19

: 00:00:35

So the first thing that we're

going to do is set up Graham's

20

: 00:00:37

profile on My Datafolio.

21

: 00:00:38

Just give a name, a portfolio URL.

22

: 00:00:40

We'll just do a headline of data analyst.

23

: 00:00:43

And for a bio, what should your bio be?

24

: 00:00:45

Something like that.

25

: 00:00:45

Looks great.

26

: 00:00:46

Data analyst with a BS in

statistics, located Provo, Utah.

27

: 00:00:50

We'll also add a quick profile

picture, which I will just steal

28

: 00:00:53

from Graham's LinkedIn even though

it's not the best photo of all time.

29

: 00:00:57

There we go.

30

: 00:00:58

What skills do you have, Graham?

31

: 00:00:59

Python, R, Excel, Pandas, Power BI.

32

: 00:01:03

All right.

33

: 00:01:04

There it is.

34

: 00:01:04

Power BI.

35

: 00:01:04

We can add some other ones, like Claude

is another one that you have used.

36

: 00:01:08

Anything else?

37

: 00:01:09

ChatGPT?

38

: 00:01:09

ChatGPT.

39

: 00:01:10

Okay.

40

: 00:01:10

Awesome.

41

: 00:01:11

Uh, we'll go ahead and link to your, uh,

GitHub profile as well and your LinkedIn

42

: 00:01:17

so that way people can contact you.

43

: 00:01:19

And, uh, we'll go ahead

and upload your resume.

44

: 00:01:22

And then what color scheme do you like?

45

: 00:01:23

Let's go with the nice

forest green right there.

46

: 00:01:25

Nice forest green.

47

: 00:01:26

We'll leave your contact

section blank for right now.

48

: 00:01:28

And do you need to do any password

protection for any of your projects?

49

: 00:01:32

I don't think so.

50

: 00:01:32

Do you have a custom

domain you'd like to use?

51

: 00:01:34

Not at this moment in time.

52

: 00:01:35

Okay, let's go ahead and hit Save Profile.

53

: 00:01:37

Okay.

54

: 00:01:37

And just like that, you have a

portfolio already made for you.

55

: 00:01:40

Boom.

56

: 00:01:41

Whoa, that's pretty cool.

57

: 00:01:42

Yeah.

58

: 00:01:42

But, uh, you'll notice this portfolio

is missing something pretty important.

59

: 00:01:46

Any work, anything.

60

: 00:01:48

Any projects, right?

61

: 00:01:49

Yeah.

62

: 00:01:49

So let's go ahead and,

uh, add some projects.

63

: 00:01:52

So when you're adding a project, there's

three different ways that we can do it.

64

: 00:01:55

The write manually, which is the way

I used to do all of my projects, um,

65

: 00:01:58

but we also have two other AI features,

which is an AI import and AI-guided form.

66

: 00:02:02

We're gonna focus today, just 'cause

we're in a time crunch, trying to do

67

: 00:02:04

this as quickly as possible, with the

AI import, which basically allows you

68

: 00:02:07

to import any sort of GitHub repos,

Tableau public links, any files

69

: 00:02:12

like Excel files or Python files or

R files you've done, and write the

70

: 00:02:15

first draft of your project for you.

71

: 00:02:18

GitHub repos are something we can try?

72

: 00:02:19

Oh, we got a couple that we can try.

73

: 00:02:21

Okay.

74

: 00:02:21

So let's go ahead and try the AI import.

75

: 00:02:22

All right.

76

: 00:02:23

This is Graham's GitHub, uh, repositories.

77

: 00:02:25

It's definitely a little bit messy.

78

: 00:02:27

Definitely maybe needs some

love, but, um, let's take a look.

79

: 00:02:29

Which one of these repos do you feel

like could be your first project?

80

: 00:02:32

Which one would be good?

81

: 00:02:32

Let's start with the non-parametic

log linear medical costs.

82

: 00:02:36

Okay.

83

: 00:02:36

You want to start here?

84

: 00:02:36

Yeah.

85

: 00:02:37

Okay.

86

: 00:02:37

Awesome.

87

: 00:02:37

And what is...

88

: 00:02:38

What exactly is this repo, I guess?

89

: 00:02:39

It's a school project that delves

into different, like, information,

90

: 00:02:43

data to quantify, uh, like how

smoking and different factors affect

91

: 00:02:48

medical costs on an annual basis.

92

: 00:02:50

Is this like a homework assignment?

93

: 00:02:51

Yes.

94

: 00:02:51

So like anyone who has done any

sort of homework assignment, this is

95

: 00:02:55

basically just a homework assignment.

96

: 00:02:56

Yes.

97

: 00:02:57

And it looks like it's in Python?

98

: 00:02:58

Yeah Okay, interesting.

99

: 00:03:00

I don't know much about this,

so we're just gonna try it.

100

: 00:03:02

So all you have to do is grab, uh, the

repo link right here, go back to our

101

: 00:03:07

AI import, go ahead and give the URL.

102

: 00:03:09

Is there any other details that we

should give it or any other instructions?

103

: 00:03:12

I don't know.

104

: 00:03:13

All right.

105

: 00:03:13

Let's just go ahead and hit

generate project article.

106

: 00:03:15

This will take a few seconds to read

through everything inside of this GitHub

107

: 00:03:18

repo and actually do the write-up.

108

: 00:03:20

All right, so it just finished

doing your project write-up here

109

: 00:03:23

and made the title Nonparametric and

Log-Linear Medical Cost Analysis.

110

: 00:03:27

That's an interesting name.

111

: 00:03:28

Okay, we're just gonna keep it as it is.

112

: 00:03:30

It gave you this URL slug.

113

: 00:03:32

It gave you this summary, "A case study

that combines nonparametric techniques

114

: 00:03:35

and log-linear modeling to predict

and interpret highly skewed medical

115

: 00:03:38

cost data, improving forecasting

robustness and interpretability."

116

: 00:03:42

Sounds pretty professional.

117

: 00:03:43

That sounds very professional.

118

: 00:03:44

And then here's an overview, the problem,

the approach, data and methodology,

119

: 00:03:48

key findings, results and impact,

conclusion, all written for you.

120

: 00:03:53

Nice.

121

: 00:03:53

Let's look at the, uh, results and impact.

122

: 00:03:55

It says more re- do you remember

anything about this project?

123

: 00:03:58

Yeah.

124

: 00:03:58

Take a look.

125

: 00:03:58

Can you read these results and impact

and see if it makes sense or not?

126

: 00:04:01

Yeah.

127

: 00:04:01

Do you want me to read them?

128

: 00:04:02

Yeah, read out loud.

129

: 00:04:03

Okay.

130

: 00:04:03

"More reliable budgeting, improved

forecasting accuracy on accurate

131

: 00:04:06

expenditures helps f- Finance teams

set reserve level with greater confi-

132

: 00:04:11

confidence Do you remember that at all?

133

: 00:04:13

Uh, yes.

134

: 00:04:14

There was like a, they had like standard

questions with like the data set for

135

: 00:04:17

the, like the presentation project, and

we, there was like findings that there

136

: 00:04:21

was like very significant correlation

between like different factors and their

137

: 00:04:26

predicted- Okay ... cost difference.

138

: 00:04:27

So it's, it's not necessarily wrong.

139

: 00:04:29

No.

140

: 00:04:29

Okay.

141

: 00:04:30

What about the...

142

: 00:04:31

Let's do this one, I guess, right here.

143

: 00:04:32

The log linear coefficients in

the two-part decomposition allowed

144

: 00:04:35

me to i- identify which variables

most strongly influence utilization

145

: 00:04:38

versus conditional costs, guiding

targeted inter- interventions.

146

: 00:04:42

That's also true because they're, uh,

like filtered out different factors and

147

: 00:04:46

variables, and I think smoking was by

far the, like most influential factor.

148

: 00:04:51

Okay.

149

: 00:04:51

So it gets some of the results right.

150

: 00:04:53

Um, I guess it also said that the log c-

of the cost was a more stable coefficient

151

: 00:04:58

and linear relationship, so the log

was the right way to do this modeling.

152

: 00:05:01

Okay, so this doesn't

feel 100% wrong to you.

153

: 00:05:03

No.

154

: 00:05:04

Okay, cool.

155

: 00:05:04

And it's telling you kind of an overview

of the project, uh, what the problem

156

: 00:05:07

is, which is predicting the medical

cost of something for like budgeting

157

: 00:05:10

reasons, and then it gives you, you

know, kind of how you did the data

158

: 00:05:13

exploration, you did the transformations,

then you did the modeling, and then

159

: 00:05:16

basically evaluate how everything went.

160

: 00:05:17

Okay, very cool.

161

: 00:05:18

So, uh, we can hit save project right here

on this project, and now if you go back to

162

: 00:05:23

your portfolio and you hit refresh, boom.

163

: 00:05:26

Whoa.

164

: 00:05:26

You got a project right here, right there.

165

: 00:05:27

That's easy.

166

: 00:05:28

All ready for you to- It's real easy.

167

: 00:05:29

That's what I like to hear.

168

: 00:05:30

Okay, awesome.

169

: 00:05:31

Let's do a- another one.

170

: 00:05:32

What is another, uh, project or

another repo that we should do?

171

: 00:05:36

Let's do the NBA heat map.

172

: 00:05:38

All right, let's do NBA

heat map right here.

173

: 00:05:40

I'm just gonna copy and paste up here.

174

: 00:05:41

Go back to add project, AI

import, paste this right here.

175

: 00:05:45

Any other instructions?

176

: 00:05:46

No, I think my .md

177

: 00:05:48

files are pretty good.

178

: 00:05:48

Okay.

179

: 00:05:49

Generate project article, and,

uh, we'll see what it does.

180

: 00:05:52

All right, it just finished.

181

: 00:05:54

NBA shot heat map explorer.

182

: 00:05:56

Let's see.

183

: 00:05:57

So I built an NBA shot heat map

explorer to turn raw NBA shot and

184

: 00:06:01

play-by-play data into actionable,

visible, intuitive insights.

185

: 00:06:04

Is that what this project's all about?

186

: 00:06:05

That's exactly what it's about.

187

: 00:06:06

Okay, let's see.

188

: 00:06:06

So then it goes through problem, the

approach, data and method- methodology.

189

: 00:06:10

So you're getting the

data from the NBA API.

190

: 00:06:13

Correct.

191

: 00:06:14

Okay.

192

: 00:06:14

And then doing some filtering,

some spatial, uh, aggregation

193

: 00:06:17

with the hex bin stuff going on.

194

: 00:06:19

Yep.

195

: 00:06:19

Okay, awesome.

196

: 00:06:20

Then you're doing some

kernel density estimators.

197

: 00:06:22

Okay, great.

198

: 00:06:23

Key findings, distinct ro- role

profiles are clearly displayed,

199

: 00:06:27

hidden inefficiencies surface

quickly, strategic match up.

200

: 00:06:30

Um, so you can do team

level heat maps to show.

201

: 00:06:32

Okay, very cool.

202

: 00:06:33

Visual artifact.

203

: 00:06:34

Very nice.

204

: 00:06:35

So, um, obviously we're just pulling

straight from the GitHub, right?

205

: 00:06:40

So it just has whatever you have- in

here, which I'm guessing it doesn't

206

: 00:06:43

have, like, any saved images, right?

207

: 00:06:46

Not in that folder particularly, no.

208

: 00:06:48

Okay.

209

: 00:06:48

See, well, that's something you could

have told me earlier when I said,

210

: 00:06:51

"Do you wanna add anything else?"

211

: 00:06:52

Well, okay, now you know.

212

: 00:06:53

We can, we can actually, like, go

in and add those images as well.

213

: 00:06:57

Um, so that would help you.

214

: 00:06:58

So let's go ahead and hit Save Project.

215

: 00:06:59

Let's go back to our portfolio, and

let's hit refresh on the full portfolio,

216

: 00:07:04

and boom, you got two projects.

217

: 00:07:06

Now, I did see that on your

LinkedIn the other day you had

218

: 00:07:09

posted about this project, right?

219

: 00:07:10

Yes.

220

: 00:07:10

So let's see.

221

: 00:07:11

Here's your LinkedIn page, and here's

the image I saw that you posted.

222

: 00:07:15

Mm-hmm.

223

: 00:07:15

I'm gonna actually right

click on this image, and I'm

224

: 00:07:17

gonna go back to our project.

225

: 00:07:20

I'm gonna go to the Heatmap Explorer here.

226

: 00:07:23

I'm gonna upload that image

as a cover image right here

227

: 00:07:26

and, uh, see how it looks.

228

: 00:07:29

Whoa.

229

: 00:07:29

Let's go back, refresh.

230

: 00:07:30

Oh, that looks way better.

231

: 00:07:31

Boom.

232

: 00:07:32

You open it up, it actually

includes that image at the top now.

233

: 00:07:34

So I like that.

234

: 00:07:35

Do we have any other images

on your LinkedIn of this?

235

: 00:07:37

No.

236

: 00:07:37

Boom.

237

: 00:07:38

Not yet.

238

: 00:07:38

Not yet.

239

: 00:07:38

Okay, sweet.

240

: 00:07:39

This is a Project and Portfolio.

241

: 00:07:40

It shows what different libraries you

used in Python and obviously Python here.

242

: 00:07:44

At the top it w- will allow people

to view your code, and you have

243

: 00:07:47

your full write-up down here.

244

: 00:07:48

Um, it allows people to see

other projects, so here's

245

: 00:07:50

your other project once again.

246

: 00:07:51

Um, here's the different

libraries you used here.

247

: 00:07:53

Um, and then you can always have

your users go back to your portfolio.

248

: 00:07:58

You can send this to people.

249

: 00:08:00

You can try dark mode or light mode.

250

: 00:08:02

It has your GitHub, your LinkedIn,

your resume, your little summary, your

251

: 00:08:06

different skills up here at the top,

your projects, and then a call to action

252

: 00:08:09

down here at the bottom to work together.

253

: 00:08:10

That's awesome.

254

: 00:08:11

Thank you so much.

255

: 00:08:11

That's gonna be very

helpful for me, I think.

256

: 00:08:13

Okay.

257

: 00:08:13

The other thing I wanted to show you

is it actually, we have these KPIs

258

: 00:08:16

here for the pro plan of MyDatafolio,

which actually shows you how many page

259

: 00:08:21

views you have and how many visitors.

260

: 00:08:22

So I'm the only person who's

visited, so it's the one.

261

: 00:08:24

So far.

262

: 00:08:24

But, like, basically it'll let you

see that this has four views, this

263

: 00:08:27

has one view, so on and so forth.

264

: 00:08:29

Um, that...

265

: 00:08:30

This is kind of exciting-

Ooh ... 'cause when someone actually

266

: 00:08:32

looks at it, you'll, you'll know.

267

: 00:08:33

Yeah, you can actually see if,

like, a recruiter or a hiring

268

: 00:08:36

person is actually looking at it.

269

: 00:08:37

Exactly.

270

: 00:08:38

So you can always edit the projects,

share a n- unique project, and share

271

: 00:08:42

your portfolio from right here.

272

: 00:08:44

That's awesome.

273

: 00:08:44

I'm excited to actually use this and get

in there and edit a few things around.

274

: 00:08:47

All right.

275

: 00:08:47

There you have it, folks.

276

: 00:08:48

I don't know how many minutes that

took, but hopefully less than 20.

277

: 00:08:51

And Graham went from having no portfolio,

just, like, some loose homework projects

278

: 00:08:56

or some projects that he's done in GitHub.

279

: 00:08:57

You can even just upload a file,

for instance, in up- add projects.

280

: 00:09:02

You can actually just upload, like,

your Python file or your Excel file

281

: 00:09:05

and it will try to do its best.

282

: 00:09:06

Obviously, the more information

you give it, the better it'll do.

283

: 00:09:09

But hopefully that gets you guys

excited to go try out MyDatafolio.com

284

: 00:09:13

and try it out for themselves.

285

: 00:09:15

Yeah, I'm excited to go and actually try

and apply to a few more jobs with this.

286

: 00:09:18

All right.

287

: 00:09:18

Link in the description.

288

: 00:09:19

Trust it out.

289

: 00:09:19

Let me know what you guys think.