WEBVTT

1
00:00:00.080 --> 00:00:04.679
<v Speaker 1>Have you ever felt just buried under repetitive tasks at

2
00:00:04.719 --> 00:00:07.879
<v Speaker 1>work or maybe just wish there was a faster I

3
00:00:07.879 --> 00:00:10.839
<v Speaker 1>don't know, a smarter way to get things done, doesn't

4
00:00:10.839 --> 00:00:12.519
<v Speaker 1>really matter what department you're in.

5
00:00:12.599 --> 00:00:14.560
<v Speaker 2>Yeah, that feeling of there has to be a better

6
00:00:14.599 --> 00:00:15.519
<v Speaker 2>way exactly.

7
00:00:15.880 --> 00:00:18.879
<v Speaker 1>If that sounds familiar, then this deep dive is definitely

8
00:00:18.960 --> 00:00:22.559
<v Speaker 1>for you. Today we're digging into automated recipes to obscue

9
00:00:22.600 --> 00:00:25.359
<v Speaker 1>your business. That a really insightful book from Packed Publishing

10
00:00:25.440 --> 00:00:28.079
<v Speaker 1>came out back in January twenty seventeen.

11
00:00:27.839 --> 00:00:31.320
<v Speaker 2>Right, and this deep dive it's all about unlocking the well,

12
00:00:31.399 --> 00:00:35.880
<v Speaker 2>the power of automation, specifically using Python. The book itself

13
00:00:35.920 --> 00:00:38.960
<v Speaker 2>is set up as a practical guide, you know, packed

14
00:00:38.960 --> 00:00:42.600
<v Speaker 2>with what they call Python recipes, basically actionable steps to

15
00:00:42.600 --> 00:00:46.039
<v Speaker 2>help businesses and well you streamline all those everyday operations

16
00:00:46.359 --> 00:00:48.560
<v Speaker 2>across like tons of different departments.

17
00:00:48.640 --> 00:00:51.200
<v Speaker 1>That's a really key point our mission here. It isn't

18
00:00:51.240 --> 00:00:53.880
<v Speaker 1>just to talk about code, right, It's more about giving

19
00:00:53.920 --> 00:00:56.799
<v Speaker 1>you the kind of shortcut a way to understand how

20
00:00:56.880 --> 00:01:00.000
<v Speaker 1>automation can spark those you know, aha moments.

21
00:00:59.719 --> 00:01:03.159
<v Speaker 2>Yeah, the ones that lead to real, actual productivity gains.

22
00:01:02.880 --> 00:01:06.519
<v Speaker 1>In your own work precisely. Yeah, and look, while the

23
00:01:06.519 --> 00:01:10.879
<v Speaker 1>book is from twenty seventeen, the core ideas the power

24
00:01:10.920 --> 00:01:14.599
<v Speaker 1>of Python for this stuff, it's still incredibly relevant today

25
00:01:14.879 --> 00:01:18.319
<v Speaker 1>for automating business processes. So let's dive in. Let's look

26
00:01:18.319 --> 00:01:20.359
<v Speaker 1>at some of the core concepts first. Okay, okay, so

27
00:01:20.439 --> 00:01:25.319
<v Speaker 1>let's unpack this central philosophy of automate it. The main

28
00:01:25.359 --> 00:01:28.359
<v Speaker 1>idea seems to be using Python to streamline well pretty

29
00:01:28.400 --> 00:01:32.799
<v Speaker 1>much everything, hr, marketing, customer support, you name it, right.

30
00:01:33.079 --> 00:01:35.840
<v Speaker 2>It's about moving away from those manual pasks, the ones

31
00:01:35.879 --> 00:01:39.640
<v Speaker 2>you do over and over, often redundantly, and shifting towards

32
00:01:39.719 --> 00:01:42.519
<v Speaker 2>more efficient, maybe even innovative business flows.

33
00:01:42.640 --> 00:01:45.120
<v Speaker 1>What's really insightful, I think, is how the book frames it.

34
00:01:45.120 --> 00:01:48.079
<v Speaker 1>It's like applying these classic problem solving patterns but to

35
00:01:48.159 --> 00:01:49.959
<v Speaker 1>all these different business challenges exactly.

36
00:01:50.280 --> 00:01:52.959
<v Speaker 2>It helps you think about automation not just as some

37
00:01:53.120 --> 00:01:56.000
<v Speaker 2>tech fix, but really as a strategic tool, a way

38
00:01:56.040 --> 00:01:57.840
<v Speaker 2>to genuinely upskill your business.

39
00:01:57.920 --> 00:02:00.400
<v Speaker 1>As the title says, right, it empowers your operation to

40
00:02:00.439 --> 00:02:03.599
<v Speaker 1>do more, but with less human.

41
00:02:03.439 --> 00:02:06.319
<v Speaker 2>Drudgery, precisely less time spent on the boring stuff.

42
00:02:06.480 --> 00:02:08.879
<v Speaker 1>Okay, so let's move into marketing and sales yeah, because

43
00:02:08.879 --> 00:02:10.599
<v Speaker 1>this is where the book it's really interesting, I think,

44
00:02:11.479 --> 00:02:15.039
<v Speaker 1>especially for anyone in those fields. Imagine the hours you

45
00:02:15.039 --> 00:02:18.560
<v Speaker 1>could save. Let's talk about lead generation first.

46
00:02:18.639 --> 00:02:18.960
<v Speaker 2>Okay.

47
00:02:19.240 --> 00:02:22.599
<v Speaker 1>It introduces this scenario with Ryan, a marketing manager at

48
00:02:22.639 --> 00:02:26.879
<v Speaker 1>a startup Deli Inc. Food delivery. Right, So Ryan needs

49
00:02:26.919 --> 00:02:29.960
<v Speaker 1>this big database of London restaurants, you know, name, address,

50
00:02:30.000 --> 00:02:34.080
<v Speaker 1>contact info, stuff to target for their platform. But searching

51
00:02:34.199 --> 00:02:36.800
<v Speaker 1>Yelp manually forget it way too slow.

52
00:02:36.919 --> 00:02:39.639
<v Speaker 2>Yeah, that's a classic problem. The solution the book offers

53
00:02:39.919 --> 00:02:43.319
<v Speaker 2>Python webscraping. Oh okay, it shows you how to automatically

54
00:02:43.319 --> 00:02:47.479
<v Speaker 2>pull out those specific details name, address, phone from Yelp

55
00:02:47.840 --> 00:02:51.120
<v Speaker 2>search results. Basically, you teach Python to look for certain

56
00:02:51.199 --> 00:02:53.719
<v Speaker 2>markers in the website's code, like the biz name or

57
00:02:53.759 --> 00:02:56.599
<v Speaker 2>street address class names they mentioned, So it just grabs

58
00:02:56.599 --> 00:02:58.879
<v Speaker 2>the data automatically pretty much. And this isn't just about

59
00:02:58.879 --> 00:03:03.360
<v Speaker 2>finding restaurants, right, It's about systematically gathering public data at scale.

60
00:03:03.719 --> 00:03:06.560
<v Speaker 2>Think about market research. And it turns that manual slog

61
00:03:06.639 --> 00:03:09.400
<v Speaker 2>into automated strategic intelligence gathering.

62
00:03:09.719 --> 00:03:12.800
<v Speaker 1>Hmmm. That opens up possibilities. And what are social media?

63
00:03:13.000 --> 00:03:15.400
<v Speaker 1>That's another huge time sink for marketing teams.

64
00:03:15.520 --> 00:03:18.560
<v Speaker 2>Oh, definitely The book uses the example of Joy, a

65
00:03:18.639 --> 00:03:22.280
<v Speaker 2>marketing head. She needs to schedule product updates across different

66
00:03:22.319 --> 00:03:25.599
<v Speaker 2>time zones, even when she's like a sleep or on vacation.

67
00:03:25.879 --> 00:03:28.439
<v Speaker 1>That's a logistical nightmare for a lot of folks.

68
00:03:28.120 --> 00:03:31.360
<v Speaker 2>Totally, So for Joy, the book shows how to use

69
00:03:31.479 --> 00:03:36.080
<v Speaker 2>Twitter's rest APIs with Python. Specifically, it mentions a tool

70
00:03:36.120 --> 00:03:38.520
<v Speaker 2>called twython. Okay, twython, which is essentially, you know, a

71
00:03:38.560 --> 00:03:42.039
<v Speaker 2>Python library that lets your program talk directly to Twitter,

72
00:03:42.159 --> 00:03:45.560
<v Speaker 2>so you can programmatically post tweets schedule them based on

73
00:03:45.639 --> 00:03:49.400
<v Speaker 2>content time and this is key time zone. It even

74
00:03:49.479 --> 00:03:52.360
<v Speaker 2>mentions using the pits module for getting those time zones right,

75
00:03:52.400 --> 00:03:53.520
<v Speaker 2>like for an Australian audience.

76
00:03:53.560 --> 00:03:55.520
<v Speaker 1>Wow, okay, So it takes social media from being this

77
00:03:55.680 --> 00:03:59.240
<v Speaker 1>constant real time burden to something you can plan out

78
00:03:59.240 --> 00:04:00.159
<v Speaker 1>and optimize.

79
00:03:59.800 --> 00:04:03.360
<v Speaker 2>Before forehand, exactly pre planned optimized outreach.

80
00:04:03.599 --> 00:04:07.520
<v Speaker 1>Now for digging even deeper, there's Judy, a data scientist

81
00:04:07.599 --> 00:04:10.520
<v Speaker 1>set a magazine. Her challenge is different. She needs to

82
00:04:10.560 --> 00:04:14.039
<v Speaker 1>collect and analyze social media data, mostly from Twitter, to

83
00:04:14.080 --> 00:04:15.560
<v Speaker 1>find insights for articles.

84
00:04:15.960 --> 00:04:19.319
<v Speaker 2>Right, So the solution there involves automating that analysis. It

85
00:04:19.399 --> 00:04:22.600
<v Speaker 2>shows using Python tools like tweepee that lets you tap

86
00:04:22.639 --> 00:04:26.560
<v Speaker 2>into Twitter's live data stream. The streaming APIs, so you're

87
00:04:26.560 --> 00:04:29.759
<v Speaker 2>getting real time tweets yep. And then for making sense

88
00:04:29.759 --> 00:04:32.319
<v Speaker 2>of it all, it brings in pandas. Think of pandas

89
00:04:32.360 --> 00:04:36.399
<v Speaker 2>as like a superpowered spreadsheet for code. It can analyze

90
00:04:36.399 --> 00:04:39.240
<v Speaker 2>the sentiment of those tweets, you know, figure out if

91
00:04:39.279 --> 00:04:42.040
<v Speaker 2>they're positive, negative, or neutral about a product like an

92
00:04:42.079 --> 00:04:43.560
<v Speaker 2>iPhone or a Samsung Note.

93
00:04:43.680 --> 00:04:45.079
<v Speaker 1>Okay, sentiment analysis.

94
00:04:45.160 --> 00:04:47.879
<v Speaker 2>Yeah, and if we connect this back to the bigger picture, right,

95
00:04:48.319 --> 00:04:51.600
<v Speaker 2>this kind of automation really shifts marketing. You move from

96
00:04:51.639 --> 00:04:54.920
<v Speaker 2>just reacting to things to being proactive using data to

97
00:04:55.040 --> 00:04:56.000
<v Speaker 2>drive your strategy.

98
00:04:56.079 --> 00:04:56.800
<v Speaker 1>Yeah, that makes sense.

99
00:04:56.920 --> 00:04:59.360
<v Speaker 2>It really raises a big question, doesn't it. How much

100
00:04:59.399 --> 00:05:02.439
<v Speaker 2>more targeted could marketing actually be if you had these

101
00:05:02.519 --> 00:05:05.319
<v Speaker 2>kinds of insights automated and like instantly available.

102
00:05:05.839 --> 00:05:09.680
<v Speaker 1>Good question. Okay, let's let's pivot now, let's talk HR

103
00:05:09.759 --> 00:05:13.040
<v Speaker 1>and ADMIN. Anyone who's dealt with onboarding new hires knows

104
00:05:13.079 --> 00:05:16.480
<v Speaker 1>the sheer volume of paperwork and repetitive steps involved.

105
00:05:16.519 --> 00:05:18.279
<v Speaker 2>Oh, absolutely, the endless forms.

106
00:05:18.519 --> 00:05:21.439
<v Speaker 1>Right. The book has this great example of an HR

107
00:05:21.480 --> 00:05:26.120
<v Speaker 1>team automating new higher orientation. I can just picture the relief. Yeah,

108
00:05:26.160 --> 00:05:29.079
<v Speaker 1>you know, dealing with fifteen twenty new hires every month.

109
00:05:29.199 --> 00:05:32.639
<v Speaker 2>Yeah, that manual process is killer. The book describes an

110
00:05:32.759 --> 00:05:36.680
<v Speaker 2>HR manager sending out personalized orientation documents based on department

111
00:05:37.040 --> 00:05:40.279
<v Speaker 2>a whole month after someone joins. Super tedious, So what's

112
00:05:40.319 --> 00:05:44.519
<v Speaker 2>a Python fix? The solution is generating personalized word documents,

113
00:05:44.839 --> 00:05:47.720
<v Speaker 2>you know, the dot dox files programmatically. It uses a

114
00:05:47.759 --> 00:05:49.839
<v Speaker 2>Python module called Python.

115
00:05:49.480 --> 00:05:51.879
<v Speaker 1>Docs Okay, python docks, right, It's.

116
00:05:51.720 --> 00:05:54.800
<v Speaker 2>A way to create and change word docs using code.

117
00:05:55.279 --> 00:05:58.160
<v Speaker 2>The script in the book pulls employee data. Okay, it

118
00:05:58.240 --> 00:06:00.399
<v Speaker 2>uses a simple dictionary in the example, but you could

119
00:06:00.439 --> 00:06:03.639
<v Speaker 2>easily hook it up to a real database and merges

120
00:06:03.680 --> 00:06:05.920
<v Speaker 2>it with an agenda template, and it just spits up

121
00:06:05.959 --> 00:06:10.120
<v Speaker 2>personalized word files exactly with the right titles, addresses department

122
00:06:10.160 --> 00:06:13.279
<v Speaker 2>specific sessions, all filled in automatically. It's a perfect case

123
00:06:13.319 --> 00:06:16.639
<v Speaker 2>study for automating something highly repetitive but also really important.

124
00:06:16.759 --> 00:06:19.360
<v Speaker 1>And it's not just onboarding, is it. I mean, think

125
00:06:19.399 --> 00:06:24.199
<v Speaker 1>about managing all sorts of office records, employee info, financial statements.

126
00:06:25.199 --> 00:06:28.560
<v Speaker 1>Doing that manually in CSVS or Excel. So much room for.

127
00:06:28.600 --> 00:06:32.399
<v Speaker 2>Error, absolutely, and so time consuming. The book gives you

128
00:06:32.519 --> 00:06:36.639
<v Speaker 2>Python recipes specifically for that simplifying and automating tasks with

129
00:06:36.720 --> 00:06:38.120
<v Speaker 2>CSV and Excel.

130
00:06:37.839 --> 00:06:40.680
<v Speaker 1>Files like what specifically, well.

131
00:06:40.519 --> 00:06:45.079
<v Speaker 2>Reading and writing CSV files, even setting up custom CSV dialects,

132
00:06:45.120 --> 00:06:49.160
<v Speaker 2>which basically means you can handle weirdly formatted data files smoothly.

133
00:06:49.759 --> 00:06:52.720
<v Speaker 2>For Excel, it covers retrieving data, putting new data, in

134
00:06:52.839 --> 00:06:56.920
<v Speaker 2>formatting cells, doing calculations with formulas, even inserting charts automatically.

135
00:06:57.040 --> 00:06:57.319
<v Speaker 1>Wow.

136
00:06:57.439 --> 00:07:01.160
<v Speaker 2>Yeah. It highlights automating income statement analysis across different years

137
00:07:01.160 --> 00:07:04.720
<v Speaker 2>for a finance team. Imagine taking hours of manual spreadsheet

138
00:07:04.720 --> 00:07:06.319
<v Speaker 2>work down to like seconds.

139
00:07:06.360 --> 00:07:10.240
<v Speaker 1>That's a huge efficiency game. Okay, let's shift again. Customer support. Yeah,

140
00:07:10.240 --> 00:07:12.800
<v Speaker 1>here it's all about being efficient, being responsive, and the

141
00:07:12.800 --> 00:07:15.360
<v Speaker 1>book has some interesting ideas here too. Take Kelly, she's

142
00:07:15.399 --> 00:07:19.399
<v Speaker 1>the director of customer support, her team, her engineers. They

143
00:07:19.399 --> 00:07:21.800
<v Speaker 1>spend way too much time on what they call level

144
00:07:21.839 --> 00:07:24.839
<v Speaker 1>one requests, you know, customers asking for info that's already

145
00:07:24.839 --> 00:07:26.439
<v Speaker 1>sitting on the FAQ page.

146
00:07:26.240 --> 00:07:29.000
<v Speaker 2>Right, and they're just manually copying and pasting links exactly.

147
00:07:29.199 --> 00:07:30.519
<v Speaker 1>So what's the automated approach?

148
00:07:30.680 --> 00:07:34.279
<v Speaker 2>The book outlines building an automated email response system. It

149
00:07:34.319 --> 00:07:37.480
<v Speaker 2>automatically acknowledges the support ticket and sends back a link

150
00:07:37.519 --> 00:07:40.600
<v Speaker 2>to the relevant FAQ section, like straight away, how does

151
00:07:40.639 --> 00:07:44.399
<v Speaker 2>it do that? It details using specific Python libraries awesomet

152
00:07:44.519 --> 00:07:48.040
<v Speaker 2>club for sending email and immaplub for fetching email, so

153
00:07:48.079 --> 00:07:50.879
<v Speaker 2>the script can check the inbox for unread support requests

154
00:07:50.920 --> 00:07:52.879
<v Speaker 2>and fire off those auto responses.

155
00:07:52.959 --> 00:07:55.560
<v Speaker 1>That frees up the support team for the harder problem.

156
00:07:55.279 --> 00:07:59.399
<v Speaker 2>Precisely reduces the load lets them focus on more complex issues.

157
00:07:59.519 --> 00:08:03.120
<v Speaker 1>The book all So talks about playing with SMS and

158
00:08:03.199 --> 00:08:06.680
<v Speaker 1>voice notifications using cloud, telefany VoIP.

159
00:08:07.000 --> 00:08:09.399
<v Speaker 2>Yeah, that opens up a whole other area for automation.

160
00:08:09.680 --> 00:08:13.240
<v Speaker 2>Think about instant updates via text or even automated voice messages.

161
00:08:13.360 --> 00:08:14.519
<v Speaker 1>What kind of examples do they give?

162
00:08:14.800 --> 00:08:18.360
<v Speaker 2>Well, things like registering with cloud to lefhany providers like

163
00:08:18.519 --> 00:08:22.639
<v Speaker 2>Twilio's a big one, sending and receiving texts automatically. It

164
00:08:22.680 --> 00:08:26.240
<v Speaker 2>even mentions SMS workflows like the ones Domino's Pizza uses

165
00:08:26.240 --> 00:08:27.120
<v Speaker 2>for order tracking.

166
00:08:27.160 --> 00:08:28.439
<v Speaker 1>Oh yeah, I get those right.

167
00:08:28.720 --> 00:08:32.159
<v Speaker 2>And beyond texts, sending automated voice messages or even building

168
00:08:32.200 --> 00:08:36.200
<v Speaker 2>parts of customer customer service software. It just extends your

169
00:08:36.200 --> 00:08:39.279
<v Speaker 2>reach and responsiveness way beyond manual limits.

170
00:08:39.360 --> 00:08:41.639
<v Speaker 1>So if we pull back a bit, what does all

171
00:08:41.679 --> 00:08:45.360
<v Speaker 1>this mean for how businesses interact with customers. More broadly,

172
00:08:45.440 --> 00:08:49.039
<v Speaker 1>I mean think about chatbots everywhere now, Pizza hut on Facebook, Messenger,

173
00:08:49.200 --> 00:08:51.120
<v Speaker 1>CNN giving you news headlines.

174
00:08:51.240 --> 00:08:52.840
<v Speaker 2>Bots are definitely a big part of it, and the

175
00:08:52.879 --> 00:08:55.559
<v Speaker 2>book explains why they're so relevant now. They arrange from

176
00:08:55.600 --> 00:08:58.279
<v Speaker 2>simple rule based ones, you know, if customer types. X

177
00:08:58.320 --> 00:09:01.000
<v Speaker 2>Spot says, why all the way to really smart AI

178
00:09:01.039 --> 00:09:02.039
<v Speaker 2>powered bots.

179
00:09:01.919 --> 00:09:03.480
<v Speaker 1>And why are they so relevant now?

180
00:09:03.679 --> 00:09:07.320
<v Speaker 2>Several reasons. One is just changing habits people spend more

181
00:09:07.320 --> 00:09:10.639
<v Speaker 2>time in chat apps. Bots are also cost effective, right,

182
00:09:11.200 --> 00:09:14.279
<v Speaker 2>reduces the need for as many human agents for simple stuff.

183
00:09:14.639 --> 00:09:17.759
<v Speaker 2>They allow for massive scale, reaching millions on platforms like

184
00:09:17.799 --> 00:09:22.200
<v Speaker 2>Facebook or Telegram, and the text got better Cheaper advances

185
00:09:22.200 --> 00:09:25.080
<v Speaker 2>in AI and natural language processing make them smarter.

186
00:09:25.559 --> 00:09:28.120
<v Speaker 1>That makes sense. Does the book get into the nitty

187
00:09:28.159 --> 00:09:30.720
<v Speaker 1>gritty of actually building these bots? Yeah, or maybe some

188
00:09:30.799 --> 00:09:31.399
<v Speaker 1>of the challenges.

189
00:09:31.480 --> 00:09:34.000
<v Speaker 2>Oh yeah, it gets practical. It shows how to build

190
00:09:34.080 --> 00:09:37.960
<v Speaker 2>Telegram bots using a library called Python telegram Bot, and

191
00:09:38.000 --> 00:09:41.080
<v Speaker 2>for Facebook Messenger it uses flask, which is a Python

192
00:09:41.120 --> 00:09:43.440
<v Speaker 2>web framework along with the Request library for talking to

193
00:09:43.440 --> 00:09:44.279
<v Speaker 2>Facebook systems.

194
00:09:44.440 --> 00:09:46.279
<v Speaker 1>Okay, so tangible tools exactly.

195
00:09:46.720 --> 00:09:49.519
<v Speaker 2>And for the smart bots, it introduces concepts like AI

196
00:09:49.639 --> 00:09:52.960
<v Speaker 2>driven sentiment analysis which we mentioned earlier, and using something

197
00:09:53.000 --> 00:09:57.279
<v Speaker 2>called AM artificial intelligence markup language. Yeah, it's a way

198
00:09:57.320 --> 00:09:59.960
<v Speaker 2>to structure the bot's knowledge so it can understand content

199
00:10:00.240 --> 00:10:03.120
<v Speaker 2>better and respond in a more well human like way.

200
00:10:03.720 --> 00:10:05.879
<v Speaker 2>The example they use is a bot for a book

201
00:10:05.879 --> 00:10:09.240
<v Speaker 2>publishing website that can have a decent conversation with customers.

202
00:10:09.240 --> 00:10:10.799
<v Speaker 2>It really moves beyond just theory.

203
00:10:11.080 --> 00:10:14.759
<v Speaker 1>Interesting, Okay. Finally, let's talk about data, making sense of

204
00:10:14.799 --> 00:10:18.080
<v Speaker 1>all the information we collect and maybe even visual information.

205
00:10:18.600 --> 00:10:22.120
<v Speaker 1>The book has this concept of imaging as a business process.

206
00:10:22.639 --> 00:10:26.120
<v Speaker 2>Right. Think about Peter who needs to digitize stacks of

207
00:10:26.159 --> 00:10:29.879
<v Speaker 2>financial documents their image based maybe scan PD apps or photos,

208
00:10:30.120 --> 00:10:32.879
<v Speaker 2>and he needs to index them. Doing that manually scanning

209
00:10:33.000 --> 00:10:36.840
<v Speaker 2>typing in data is slow, expensive, error.

210
00:10:36.600 --> 00:10:39.120
<v Speaker 1>Prone definitely, So how does automation help there?

211
00:10:39.200 --> 00:10:42.919
<v Speaker 2>The solution involves automating both the spanning and the indexing

212
00:10:43.039 --> 00:10:47.279
<v Speaker 2>using Python. It leverages powerful libraries like OpenCV and psychic

213
00:10:47.320 --> 00:10:50.799
<v Speaker 2>image for image processing stuff like finding the document edges

214
00:10:50.799 --> 00:10:53.519
<v Speaker 2>in a photo okay, and then it uses ocr tools

215
00:10:53.559 --> 00:10:56.559
<v Speaker 2>optical character recognition like tesseract and its Python wrapper py

216
00:10:56.600 --> 00:10:59.840
<v Speaker 2>Tesseract basically teaching the computer to read the text with

217
00:10:59.879 --> 00:11:01.200
<v Speaker 2>it the image, so it.

218
00:11:01.159 --> 00:11:05.039
<v Speaker 1>Can turn a picture of a document into actual text exactly.

219
00:11:05.480 --> 00:11:08.320
<v Speaker 2>The book shows a recipe take an image, say a

220
00:11:08.360 --> 00:11:12.879
<v Speaker 2>newspaper clipping, detect its edges, identify the text areas, digitally

221
00:11:12.960 --> 00:11:15.399
<v Speaker 2>scan it to make it clean black and white, and

222
00:11:15.440 --> 00:11:18.200
<v Speaker 2>then extract the actual text content.

223
00:11:18.559 --> 00:11:18.840
<v Speaker 1>Wow.

224
00:11:19.279 --> 00:11:22.440
<v Speaker 2>Yeah, and this really brings up a key question about efficiency. Right,

225
00:11:22.919 --> 00:11:27.080
<v Speaker 2>imagine going from boxes of paper records to a fully indexed,

226
00:11:27.360 --> 00:11:33.240
<v Speaker 2>instantly searchable digital archive for compliance for research, all automated.

227
00:11:33.600 --> 00:11:35.480
<v Speaker 2>What stands out to you as the biggest win there?

228
00:11:35.720 --> 00:11:37.519
<v Speaker 1>Oh, that's a good win for me. I think it's

229
00:11:37.559 --> 00:11:40.399
<v Speaker 1>exactly that compliance and archives taking that massive headache of

230
00:11:40.519 --> 00:11:43.799
<v Speaker 1>finding specific paper documents and turning it into like a

231
00:11:44.000 --> 00:11:46.879
<v Speaker 1>simple keyword search. That time save for audits or just

232
00:11:46.919 --> 00:11:48.799
<v Speaker 1>finding old information seems enormous.

233
00:11:48.919 --> 00:11:51.320
<v Speaker 2>Totally agree, speed and accessibility, And.

234
00:11:51.360 --> 00:11:53.080
<v Speaker 1>You know, speaking of data, the book doesn't just stop

235
00:11:53.080 --> 00:11:55.519
<v Speaker 1>at images. It lays out a solid process for general

236
00:11:55.559 --> 00:11:58.480
<v Speaker 1>data analysis and visualization too, the steps you need for

237
00:11:58.519 --> 00:11:59.960
<v Speaker 1>making decisions based on data.

238
00:12:00.159 --> 00:12:03.279
<v Speaker 2>Yes, it covers the whole workflow, starting with Okay, what's

239
00:12:03.279 --> 00:12:06.399
<v Speaker 2>your hypothesis? What question are you trying to answer? Then

240
00:12:06.519 --> 00:12:09.679
<v Speaker 2>finding the data sources, collecting it, cleaning it up which

241
00:12:09.720 --> 00:12:14.080
<v Speaker 2>is super important, removing duplicates, dealing with weird outliers.

242
00:12:14.200 --> 00:12:15.720
<v Speaker 1>Right, the data wrangling part.

243
00:12:15.720 --> 00:12:19.320
<v Speaker 2>Exactly, then the actual analysis and finally visualizing it so

244
00:12:19.320 --> 00:12:20.480
<v Speaker 2>people can understand it.

245
00:12:20.960 --> 00:12:22.519
<v Speaker 1>What tools does it recommend for that?

246
00:12:22.960 --> 00:12:29.519
<v Speaker 2>Python tools naturally, pandas again for handling data and tables, filtering, summarizing,

247
00:12:30.399 --> 00:12:34.000
<v Speaker 2>NumPy for heavy duty math, and then for creating charts

248
00:12:34.039 --> 00:12:38.159
<v Speaker 2>and graphs matplotlib and seaborn, and the use cases things

249
00:12:38.240 --> 00:12:41.200
<v Speaker 2>like reading and interpreting data. It uses an example of

250
00:12:41.279 --> 00:12:44.159
<v Speaker 2>tech Crunch funding data, looking at it by investment category,

251
00:12:44.200 --> 00:12:48.120
<v Speaker 2>funding round city, generating insights by filtering and aggregating like

252
00:12:48.320 --> 00:12:50.919
<v Speaker 2>which funding rounds were most common each month. And as

253
00:12:50.919 --> 00:12:53.720
<v Speaker 2>we talked about with Judy, automating social media analysis fits

254
00:12:53.759 --> 00:12:56.000
<v Speaker 2>here too. It really gives you a blueprint for becoming

255
00:12:56.000 --> 00:12:56.840
<v Speaker 2>more data driven.

256
00:12:57.200 --> 00:12:59.600
<v Speaker 1>Okay. And just to tie up the administrative side, the

257
00:12:59.639 --> 00:13:03.120
<v Speaker 1>book tackles time in the zone, specifically for things like

258
00:13:03.240 --> 00:13:04.720
<v Speaker 1>automatic invoice generation.

259
00:13:05.159 --> 00:13:09.240
<v Speaker 2>Right, because businesses always struggle with dates and times, especially

260
00:13:09.240 --> 00:13:14.039
<v Speaker 2>across time zones, for scheduling reports financial stuff like invoicing,

261
00:13:14.720 --> 00:13:15.720
<v Speaker 2>it gets messy fast.

262
00:13:15.759 --> 00:13:18.039
<v Speaker 1>Yeah, deal, I saving leap years exactly.

263
00:13:18.440 --> 00:13:21.279
<v Speaker 2>So. The solution uses pythons built in date time and

264
00:13:21.320 --> 00:13:24.840
<v Speaker 2>calendar modules, plus that PET library again for handling time

265
00:13:24.919 --> 00:13:28.879
<v Speaker 2>zones properly. The book gives a recipe for generating personalized

266
00:13:28.919 --> 00:13:33.080
<v Speaker 2>invoices automatically, like for a specific customer YEP, using customer

267
00:13:33.159 --> 00:13:36.639
<v Speaker 2>ID name the billing month, and it handles all the

268
00:13:36.759 --> 00:13:39.960
<v Speaker 2>date math correctly, like figuring out past dates even across

269
00:13:40.039 --> 00:13:44.440
<v Speaker 2>leap years ensures accuracy timeliness, even if you're dealing with

270
00:13:44.480 --> 00:13:45.600
<v Speaker 2>clients worldwide.

271
00:13:46.000 --> 00:13:48.679
<v Speaker 1>Okay, wow, so we've really only scratched the surface here,

272
00:13:48.720 --> 00:13:50.759
<v Speaker 1>but you get a sense of the breadth of automating,

273
00:13:51.080 --> 00:13:55.360
<v Speaker 1>generating leads, scheduling tweets, onboarding new hires, handling customer support,

274
00:13:55.639 --> 00:13:58.000
<v Speaker 1>digitizing documents, analyzing.

275
00:13:57.559 --> 00:14:00.919
<v Speaker 2>Data, and it's all powered by these Python rescids. This

276
00:14:01.039 --> 00:14:03.279
<v Speaker 2>deep dive, it really does offer you a shortcut, a

277
00:14:03.320 --> 00:14:06.279
<v Speaker 2>way to grasp how these practical techniques can solve actual

278
00:14:06.360 --> 00:14:09.759
<v Speaker 2>business problems and hopefully spark some innovation in your own work.

279
00:14:09.919 --> 00:14:11.759
<v Speaker 1>Yeah, it gives you powerful tools and helps you see

280
00:14:11.759 --> 00:14:15.240
<v Speaker 1>where those efficiencies might be hiding in your own processes. Absolutely,

281
00:14:15.360 --> 00:14:18.360
<v Speaker 1>so here's something to think about as we wrap up

282
00:14:19.639 --> 00:14:23.440
<v Speaker 1>If businesses can automate so much of the how repetitive tasks,

283
00:14:23.519 --> 00:14:26.440
<v Speaker 1>the data processing, What new frontiers and what does that

284
00:14:26.519 --> 00:14:28.840
<v Speaker 1>open up? What becomes possible when we can focus more

285
00:14:28.919 --> 00:14:32.200
<v Speaker 1>human brain power on strategy and creativity instead of just

286
00:14:32.200 --> 00:14:35.440
<v Speaker 1>the execution. How might that truly transform the future of work,