Meeting Title: Casie <> Hannah - Insomnia Cookies Date: 2025-10-20 Meeting participants: Casie Aviles, Hannah Wang
WEBVTT
1 00:01:48.300 ⇒ 00:02:03.219 Hannah Wang: Hi, Casey. Hi. I’m sorry if we can hear the voices in the back. I’m co-working with Robert in the same room, so let me know if you can’t hear me, because I’m trying to whisper, because he’s in a meeting, too.
2 00:02:03.220 ⇒ 00:02:06.519 Casie Aviles: Oh, sure, yeah. Yeah, so…
3 00:02:06.570 ⇒ 00:02:23.060 Hannah Wang: Apologies in advance, but, okay. We’ve done many case studies together, so it’ll be the same thing. Okay. And this will be for, insomnia, for browser automation.
4 00:02:23.160 ⇒ 00:02:35.410 Hannah Wang: For data scraping. So, I’m just gonna start asking you a bunch of questions, and you’ll answer them for me. So…
5 00:02:35.650 ⇒ 00:02:38.090 Hannah Wang: Give me one second.
6 00:02:38.300 ⇒ 00:02:44.229 Hannah Wang: So, can you just give me, like, a high-level overview of what this project is?
7 00:02:45.160 ⇒ 00:02:48.540 Casie Aviles: Yeah, sure, so for this project.
8 00:02:49.060 ⇒ 00:02:52.580 Casie Aviles: Basically, the goal was to automate
9 00:02:53.490 ⇒ 00:02:59.450 Casie Aviles: like, the manual processes that, the client, Insomnia Cookies, does, like.
10 00:02:59.860 ⇒ 00:03:03.560 Casie Aviles: So what they do is… They…
11 00:03:04.210 ⇒ 00:03:09.049 Casie Aviles: they manually fill out these, trackers, so they have… I’ll share my screen, so…
12 00:03:09.440 ⇒ 00:03:10.150 Hannah Wang: Okay.
13 00:03:14.090 ⇒ 00:03:14.680 Casie Aviles: Alright.
14 00:03:16.960 ⇒ 00:03:18.489 Casie Aviles: Let me know if you can see it now.
15 00:03:20.500 ⇒ 00:03:21.170 Hannah Wang: Yes.
16 00:03:22.060 ⇒ 00:03:22.700 Casie Aviles: Okay.
17 00:03:22.850 ⇒ 00:03:31.570 Casie Aviles: So, just a brief overview, so they have a bunch of trackers, so they have this daily scorecard, they have a marketing
18 00:03:32.070 ⇒ 00:03:37.459 Casie Aviles: Tracker, and then they also have one for their you know, FDA…
19 00:03:37.750 ⇒ 00:03:43.750 Casie Aviles: stuff, so for Uber and DoorDash, so these are, like, the three main trackers that
20 00:03:44.000 ⇒ 00:03:46.260 Casie Aviles: We… they manually update, so…
21 00:03:46.610 ⇒ 00:03:53.320 Casie Aviles: What they would usually do… what they typically would do is they would go to, to the web…
22 00:03:53.790 ⇒ 00:03:56.689 Casie Aviles: Platforms, and then they would just copy-paste.
23 00:03:58.040 ⇒ 00:04:03.980 Casie Aviles: So, for example, they would go to the Uber, and then they would copy-paste, like, the…
24 00:04:04.250 ⇒ 00:04:07.090 Casie Aviles: The metrics, you know, the performance of their,
25 00:04:07.900 ⇒ 00:04:10.970 Casie Aviles: Of their ads, and then their promotions.
26 00:04:12.070 ⇒ 00:04:14.730 Casie Aviles: Which takes up a lot of time, and…
27 00:04:15.090 ⇒ 00:04:16.940 Casie Aviles: They have to do that, like.
28 00:04:17.470 ⇒ 00:04:21.779 Casie Aviles: They have… they have to also backfill every… for, like.
29 00:04:22.130 ⇒ 00:04:27.249 Casie Aviles: You know, like, a 7-day window for that, so there’s a lot of manual stuff happening.
30 00:04:27.410 ⇒ 00:04:28.830 Hannah Wang: Yeah. And…
31 00:04:30.370 ⇒ 00:04:36.470 Casie Aviles: Where we come in is, basically, we just want to cut down the time that it takes,
32 00:04:36.940 ⇒ 00:04:39.909 Casie Aviles: And automate that and optimize.
33 00:04:40.080 ⇒ 00:04:42.169 Hannah Wang: the process, so…
34 00:04:42.210 ⇒ 00:04:47.750 Casie Aviles: That’s where we… that’s where we come in, and then we built, basically,
35 00:04:48.290 ⇒ 00:04:53.400 Casie Aviles: Automations around this, so we have, like, a browser automation that lives in…
36 00:04:54.830 ⇒ 00:05:00.709 Casie Aviles: yeah, browser-based, and we have a bunch of tools that we use, but yeah, that’s… I think that’s the gist of.
37 00:05:01.220 ⇒ 00:05:01.690 Hannah Wang: Okay.
38 00:05:01.690 ⇒ 00:05:04.039 Casie Aviles: The work that we have, you know.
39 00:05:04.590 ⇒ 00:05:08.630 Hannah Wang: Okay, and how long did this project take us?
40 00:05:09.980 ⇒ 00:05:16.610 Casie Aviles: Around, month, we were… so we started around August, I think second week, August.
41 00:05:17.060 ⇒ 00:05:20.849 Casie Aviles: Okay. And then we started building out the automations.
42 00:05:20.850 ⇒ 00:05:21.280 Hannah Wang: Credit.
43 00:05:21.280 ⇒ 00:05:24.449 Casie Aviles: And then around… the third week…
44 00:05:25.150 ⇒ 00:05:29.020 Casie Aviles: of September, we were wrapping up with the automations.
45 00:05:29.270 ⇒ 00:05:33.590 Casie Aviles: And right now, we’re still kind of maintaining this
46 00:05:34.080 ⇒ 00:05:42.120 Casie Aviles: Process, but it… we’re not, like, doing any heavy development on Any new automations, but yeah.
47 00:05:42.120 ⇒ 00:05:52.890 Hannah Wang: That’s pretty much the timeline. Okay. And then, you were the lead engineer. Who’s the PM for this project?
48 00:05:53.690 ⇒ 00:05:56.110 Casie Aviles: Initially, it was Amber.
49 00:05:56.510 ⇒ 00:05:59.430 Casie Aviles: And then… Justin came on.
50 00:05:59.920 ⇒ 00:06:00.640 Hannah Wang: Okay.
51 00:06:00.880 ⇒ 00:06:01.270 Casie Aviles: Yeah.
52 00:06:01.270 ⇒ 00:06:03.600 Hannah Wang: Okay, I’m just gonna put Amber, because…
53 00:06:03.790 ⇒ 00:06:08.679 Hannah Wang: I don’t… I don’t think Justin’s with us anymore, so… Yeah. Okay,
54 00:06:08.940 ⇒ 00:06:11.480 Hannah Wang: Cool, so you kind of gave…
55 00:06:12.070 ⇒ 00:06:24.899 Hannah Wang: I know you gave, like, an overview, but… so bear with me if I ask questions and make you repeat things, but the context is, like, the working environment. Before the project, everything was manual, and…
56 00:06:24.900 ⇒ 00:06:25.560 Casie Aviles: Yes.
57 00:06:25.560 ⇒ 00:06:38.229 Hannah Wang: their team had to just backfill everything and also manually fill in current data, within 3 different spreadsheets, essentially, and… or 3 different platforms, I guess.
58 00:06:38.350 ⇒ 00:06:49.779 Hannah Wang: And I guess, were there any other, like, previous efforts to automate this, or did we come in as, like, the first time on automating this entire process?
59 00:06:50.780 ⇒ 00:06:54.419 Casie Aviles: Oh, regarding that, I, I don’t…
60 00:06:54.760 ⇒ 00:07:00.729 Casie Aviles: have any, like, I don’t think they have any, like, Attempts to automate this.
61 00:07:00.730 ⇒ 00:07:01.500 Hannah Wang: Okay.
62 00:07:02.170 ⇒ 00:07:06.600 Casie Aviles: at least I haven’t heard that they, you know, that they attempted any…
63 00:07:07.070 ⇒ 00:07:10.709 Casie Aviles: solutions for that, so I think that’s very… Okay. Yeah.
64 00:07:11.780 ⇒ 00:07:22.340 Hannah Wang: And then, for… these… like, sheets? Is it just to track their sales and, like, revenue, or…
65 00:07:22.930 ⇒ 00:07:25.680 Hannah Wang: Is it something else?
66 00:07:25.680 ⇒ 00:07:32.149 Casie Aviles: Yeah, like, these are the… just, you know, they compare, so I’m not… I’m not super…
67 00:07:33.230 ⇒ 00:07:36.939 Casie Aviles: I guess I don’t know too much about, like, the marketing stuff, but…
68 00:07:36.940 ⇒ 00:07:37.690 Hannah Wang: Yeah.
69 00:07:37.690 ⇒ 00:07:47.149 Casie Aviles: Basically, what they’re just trying to do here is they’re trying to compare, like, the cost… they’re getting the cost and then the revenue.
70 00:07:47.460 ⇒ 00:07:55.339 Casie Aviles: And then they’re comparing that from, like, the current year, they’re getting the data for the current year, and then comparing that against the past year.
71 00:07:55.340 ⇒ 00:08:04.169 Hannah Wang: Oh, yep, I see that. Okay. Yeah, just comparing revenue and marketing and stuff like that. Okay.
72 00:08:04.390 ⇒ 00:08:21.110 Hannah Wang: And then the next section is to talk about the challenges, but I feel like you already mentioned that, where it was manual, and it probably took them a long time to update. I’m assuming there’s, like, human error as well, because it’s manual. Yes. And then,
73 00:08:22.500 ⇒ 00:08:31.260 Hannah Wang: I guess… yeah, and, like, I’m sure mistakes and, like, longer time-taken impacts,
74 00:08:32.090 ⇒ 00:08:39.520 Hannah Wang: their ability to adjust based on the data, and I’m assuming it’s just, like, frustrating for the business as a whole.
75 00:08:39.850 ⇒ 00:08:40.559 Casie Aviles: Armiah.
76 00:08:41.750 ⇒ 00:08:47.940 Hannah Wang: And… Okay, yeah, I feel like that the challenge is straightforward. Yeah, go ahead.
77 00:08:47.940 ⇒ 00:08:54.420 Casie Aviles: I think roughly… roughly before, like, with their manual process, it takes them around 2 hours, and…
78 00:08:55.450 ⇒ 00:09:09.290 Casie Aviles: Right now, with our automations, we were able to cut it down to around 30 minutes, although, yeah, there are… it’s not, like, completely automated, because there are a lot of, you know, access issues as well, and…
79 00:09:09.290 ⇒ 00:09:10.929 Hannah Wang: Not a lot of, like…
80 00:09:11.140 ⇒ 00:09:19.609 Casie Aviles: the data sources that we have are not, like, the best for automating, so sometimes we don’t have, like, APIs for…
81 00:09:20.290 ⇒ 00:09:25.710 Casie Aviles: Some of the data sources, so that’s where the browser automation comes in.
82 00:09:26.050 ⇒ 00:09:34.240 Hannah Wang: Okay. Yeah, so how about you just walk me through the solution, just the whole flow, including all the tools and everything?
83 00:09:34.960 ⇒ 00:09:39.329 Casie Aviles: Yeah, sure, so, yeah, like, these are the… these…
84 00:09:39.690 ⇒ 00:09:45.859 Casie Aviles: three spreadsheets here are, like, the main… the destination.
85 00:09:46.270 ⇒ 00:09:50.080 Casie Aviles: And We did build… we did have, like, a…
86 00:09:50.290 ⇒ 00:09:53.709 Casie Aviles: an intermediate, what do you call this? It’s like…
87 00:09:54.260 ⇒ 00:09:59.620 Casie Aviles: This is, like, our staging sheets, where we, where we have our…
88 00:09:59.790 ⇒ 00:10:04.149 Casie Aviles: This is where the automation basically dumps the data that…
89 00:10:04.290 ⇒ 00:10:06.560 Casie Aviles: It has, that we’ve scraped.
90 00:10:06.820 ⇒ 00:10:13.349 Casie Aviles: Where we’ve extracted… And basically, it’s like, you know, it’s just a copy of the existing
91 00:10:13.750 ⇒ 00:10:15.510 Casie Aviles: Format that they have.
92 00:10:15.710 ⇒ 00:10:17.240 Hannah Wang: Yep. And then you…
93 00:10:17.680 ⇒ 00:10:20.169 Casie Aviles: You can see here, like, we have,
94 00:10:20.330 ⇒ 00:10:24.949 Casie Aviles: Individual sheets for, like, each data source, so we have…
95 00:10:25.240 ⇒ 00:10:28.359 Casie Aviles: So they have this platform called Braze, where…
96 00:10:28.540 ⇒ 00:10:30.810 Casie Aviles: I think this is where they have, like.
97 00:10:31.030 ⇒ 00:10:35.050 Casie Aviles: where they deploy their campaigns, like their email, SMS.
98 00:10:36.650 ⇒ 00:10:45.830 Casie Aviles: Different channels, so this is their own marketing… And then… basically, what I did…
99 00:10:46.020 ⇒ 00:10:49.770 Casie Aviles: Here, what we did here is we have a bunch of automation, so…
100 00:10:50.250 ⇒ 00:10:59.729 Casie Aviles: for example, this one… this one lives on Dagster, so this is… Dagster is responsible for, like, orchestrating and scheduling these scripts.
101 00:11:00.260 ⇒ 00:11:01.869 Casie Aviles: to run.
102 00:11:03.050 ⇒ 00:11:07.579 Casie Aviles: So, yeah, this is, these are scripts in Python, and…
103 00:11:08.360 ⇒ 00:11:19.820 Casie Aviles: These are… yeah, these are basically responsible for getting the data and putting that… writing that into Google Sheets. So it’s a bunch of connectors to the data source and then to the destination.
104 00:11:20.080 ⇒ 00:11:24.609 Casie Aviles: So we have this one, Braze Campaigns Job, and then we have DoorDash.
105 00:11:25.480 ⇒ 00:11:35.799 Casie Aviles: Yeah, so… other tools that we use for the browser automation, we use BrowserBase.
106 00:11:36.200 ⇒ 00:11:39.730 Casie Aviles: So, as you can see here, so this is, like, a recording that
107 00:11:40.750 ⇒ 00:11:45.759 Casie Aviles: it captured, but essentially what browser base does is…
108 00:11:46.290 ⇒ 00:11:51.560 Casie Aviles: It handles, like, the… the… it basically mimics, like, the manual process that…
109 00:11:52.210 ⇒ 00:11:56.639 Casie Aviles: One would take to do… to do all of that, so…
110 00:11:57.990 ⇒ 00:12:05.930 Casie Aviles: you know, it would go to… it would go and visit the website, the URL, it would log in.
111 00:12:06.400 ⇒ 00:12:11.020 Casie Aviles: And then it would, you know, it would basically just copy and paste as well, but…
112 00:12:11.880 ⇒ 00:12:16.880 Casie Aviles: Instead of, you know, a human doing it, it’s going to be this browser automation, so…
113 00:12:18.250 ⇒ 00:12:20.140 Casie Aviles: Yeah, I think, and then…
114 00:12:20.360 ⇒ 00:12:26.939 Casie Aviles: it’s a little… it’s not just on Dogster, like some… like I mentioned, it’s, like, the data source is not…
115 00:12:27.780 ⇒ 00:12:34.489 Casie Aviles: The cleanest, like, Some of it works with Dagster, some of it we have… we use…
116 00:12:35.260 ⇒ 00:12:37.870 Casie Aviles: other tools, like Polyatomic.
117 00:12:37.880 ⇒ 00:12:39.200 Hannah Wang: Okay.
118 00:12:39.550 ⇒ 00:12:44.640 Casie Aviles: And one of the things there is for Google Ads, they also have, like, paid media.
119 00:12:45.100 ⇒ 00:12:48.440 Casie Aviles: That we tracked, so…
120 00:12:48.550 ⇒ 00:12:58.189 Casie Aviles: you know, the data that is from Google Ads and Meta, you know, for Google Ads, we have it… we get it using Polyatomic.
121 00:12:58.840 ⇒ 00:13:03.670 Casie Aviles: And then it also writes it here to this same staging sheet that we have.
122 00:13:04.190 ⇒ 00:13:09.080 Casie Aviles: So I think that, yeah, that’s pretty much how we built it, like, the main…
123 00:13:09.500 ⇒ 00:13:16.409 Casie Aviles: Browser automation tool that we use is browser-based, and then Dugster for orchestration.
124 00:13:17.890 ⇒ 00:13:22.099 Hannah Wang: And then how… from the staging website, does it go to the first
125 00:13:22.290 ⇒ 00:13:25.250 Hannah Wang: Like, the 3 that you showed me in the beginning?
126 00:13:27.120 ⇒ 00:13:27.989 Casie Aviles: Oh, you mean…
127 00:13:27.990 ⇒ 00:13:37.120 Hannah Wang: Or does the team look at the staging? Like, does the Insomnia team look at the staging sheet, or does it, like, copy over to the first three that you showed me?
128 00:13:37.590 ⇒ 00:13:43.450 Casie Aviles: Yeah, for that, we have to do it, we have to copy that manually here.
129 00:13:43.450 ⇒ 00:13:44.680 Hannah Wang: Oh, okay.
130 00:13:44.890 ⇒ 00:13:46.320 Hannah Wang: So yeah, it’s not…
131 00:13:46.370 ⇒ 00:13:50.070 Casie Aviles: Yeah, I did mention it wasn’t completely automated, but…
132 00:13:50.070 ⇒ 00:13:50.630 Hannah Wang: Yep.
133 00:13:51.480 ⇒ 00:13:54.930 Casie Aviles: Yeah, I think, yeah, that’s how it works right now.
134 00:13:56.060 ⇒ 00:13:57.150 Casie Aviles: And there are also…
135 00:13:57.150 ⇒ 00:13:58.429 Hannah Wang: Currently… oh, go ahead.
136 00:13:58.430 ⇒ 00:14:04.620 Casie Aviles: Yeah. Yeah, there were also a bunch of other steps that we couldn’t automate yet, because,
137 00:14:05.140 ⇒ 00:14:09.349 Casie Aviles: Yeah, like, technically, it’s not very feasible, or, like…
138 00:14:10.140 ⇒ 00:14:17.950 Casie Aviles: Yeah, I think that’s why it’s a mix of manual and… automated.
139 00:14:18.390 ⇒ 00:14:29.920 Hannah Wang: I see. And then are we… are we currently, like, running the scripts and copying it over to the spreadsheet, or did we hand that off to their team?
140 00:14:31.610 ⇒ 00:14:41.360 Casie Aviles: No, no, we’re still doing that. I mean, we’re still doing… transferring, like, you know, we’re handling… getting the data from the staging sheet to their…
141 00:14:41.360 ⇒ 00:14:41.880 Hannah Wang: Yeah.
142 00:14:41.880 ⇒ 00:14:44.319 Casie Aviles: Actual sheet, so we’re doing that.
143 00:14:45.100 ⇒ 00:14:52.350 Casie Aviles: But the scripts are, you know, scheduled to run, like, every 8 AM Eastern Time.
144 00:14:53.720 ⇒ 00:14:54.640 Hannah Wang: Every day.
145 00:14:54.890 ⇒ 00:14:55.570 Casie Aviles: Yes.
146 00:14:56.090 ⇒ 00:14:56.820 Hannah Wang: Okay.
147 00:14:59.340 ⇒ 00:15:10.899 Hannah Wang: Okay, so the tools that you used, just reiterating, Dagster, BrowserBase, Polytomic for some of them, obviously Google Sheets, I guess, and…
148 00:15:10.900 ⇒ 00:15:11.260 Casie Aviles: Yes.
149 00:15:11.260 ⇒ 00:15:15.190 Hannah Wang: Braze? Are we using Braze?
150 00:15:15.660 ⇒ 00:15:16.450 Hannah Wang: Alright.
151 00:15:16.450 ⇒ 00:15:21.810 Casie Aviles: It’s not really a tool, it’s more of, like, their platform, where they… I see.
152 00:15:23.340 ⇒ 00:15:25.860 Hannah Wang: Okay, so it’s just those 4 tools.
153 00:15:26.590 ⇒ 00:15:27.250 Casie Aviles: Yeah.
154 00:15:27.720 ⇒ 00:15:29.700 Hannah Wang: Okay,
155 00:15:30.010 ⇒ 00:15:43.590 Hannah Wang: And then, I know you mentioned this before, like, it takes… it took their team around 2 hours, but now it takes around 30 minutes, so that’s… that’s a good metric I can use. Any other metrics or feedback that we…
156 00:15:43.740 ⇒ 00:15:49.429 Hannah Wang: general feedback that we got from the Insomnia team about this automation.
157 00:15:51.610 ⇒ 00:15:58.870 Casie Aviles: I don’t think they have a lot of feedback there. It’s mostly just…
158 00:16:00.160 ⇒ 00:16:12.699 Casie Aviles: Because, you know, they basically just handed it over to us, but I think their feedback was mostly just, you know, checking, like, if something’s been updated, why something looks off, you know?
159 00:16:12.700 ⇒ 00:16:13.540 Hannah Wang: Hmm.
160 00:16:13.960 ⇒ 00:16:17.810 Casie Aviles: But, yeah, I think definitely there’s, like…
161 00:16:18.760 ⇒ 00:16:21.489 Casie Aviles: what do you call this? There’s, like, less…
162 00:16:22.670 ⇒ 00:16:27.379 Casie Aviles: sheets that we have to go through, like, before, there were even more sheets, like.
163 00:16:27.800 ⇒ 00:16:33.830 Casie Aviles: Like, these are the 3 main ones, but we have even more sources like…
164 00:16:34.210 ⇒ 00:16:36.769 Casie Aviles: For this section, we have to…
165 00:16:37.560 ⇒ 00:16:44.810 Casie Aviles: Yeah, there’s just a lot of sources, but I think we were able to cut it down, so now we’re just looking at just this one.
166 00:16:45.760 ⇒ 00:16:47.540 Casie Aviles: And we’re just copying it.
167 00:16:47.890 ⇒ 00:16:48.680 Casie Aviles: Basically.
168 00:16:48.680 ⇒ 00:16:49.510 Hannah Wang: I see.
169 00:16:50.340 ⇒ 00:16:53.720 Hannah Wang: How did you get the initial list of,
170 00:16:54.950 ⇒ 00:17:09.779 Hannah Wang: like, Grubhub, DoorDash? Like, did they just give you those delivery service, like, where they sell their product, and then you just, like, scrape the information from there? Is that how they kind of handed it off to you? Like, Uber, DoorDash?
171 00:17:09.920 ⇒ 00:17:10.550 Hannah Wang: Hmm.
172 00:17:11.170 ⇒ 00:17:14.030 Casie Aviles: They just… basically, I just,
173 00:17:15.109 ⇒ 00:17:18.790 Casie Aviles: I just got, like, the login that they provided.
174 00:17:18.790 ⇒ 00:17:20.150 Hannah Wang: to raw burn.
175 00:17:20.150 ⇒ 00:17:21.219 Casie Aviles: And then… Okay.
176 00:17:21.540 ⇒ 00:17:27.229 Casie Aviles: Yeah, I basically just, worked… started working on that as soon as I got access.
177 00:17:28.680 ⇒ 00:17:30.720 Hannah Wang: Gotcha. And then…
178 00:17:31.120 ⇒ 00:17:39.480 Hannah Wang: for browser-based, could you explain to me what that is again? Like, it… does it… it mimics, like, a…
179 00:17:39.650 ⇒ 00:17:43.150 Hannah Wang: A session of you, like, going through the.
180 00:17:43.150 ⇒ 00:17:43.790 Casie Aviles: campaigns or whatever.
181 00:17:43.790 ⇒ 00:17:45.239 Hannah Wang: Whatever. Okay.
182 00:17:46.090 ⇒ 00:17:50.999 Hannah Wang: And then how does that… how do you, like, automate stuff using that recording?
183 00:17:52.330 ⇒ 00:18:00.690 Casie Aviles: Yeah, so… basically, I’ll try to explain this as well as I can, but…
184 00:18:00.870 ⇒ 00:18:05.969 Casie Aviles: So we have, like, a Python… we have a Python script that
185 00:18:06.160 ⇒ 00:18:12.550 Casie Aviles: Basically, imports browser-based as, like a library, something like that, and then…
186 00:18:13.470 ⇒ 00:18:20.549 Casie Aviles: We were able to use, like, we were able to create, like, a session, like, a headless browser in that script.
187 00:18:22.230 ⇒ 00:18:24.600 Casie Aviles: And then that’s also where we could…
188 00:18:25.630 ⇒ 00:18:35.260 Casie Aviles: I mean, it… I’m not… there’s, like, code for that, but… Very nice.
189 00:18:36.560 ⇒ 00:18:40.870 Casie Aviles: I’m not sure if I should, I should show the code, but…
190 00:18:41.600 ⇒ 00:18:44.140 Hannah Wang: That’s okay, you don’t have to show the code.
191 00:18:44.140 ⇒ 00:18:44.850 Casie Aviles: Yeah,
192 00:18:45.520 ⇒ 00:18:54.120 Casie Aviles: But yeah, basically, it’s, like, the Dagster, the Python script, and Dagster is responsible for, for running those.
193 00:18:55.050 ⇒ 00:18:57.890 Casie Aviles: Brezerbase is just, like, a tool for us to…
194 00:18:58.230 ⇒ 00:19:01.869 Casie Aviles: Make it easier to, to, to create, like.
195 00:19:02.250 ⇒ 00:19:10.410 Casie Aviles: Headless browser, so it’s like, you know, it’s kind of similar to having another window opened And then…
196 00:19:10.410 ⇒ 00:19:11.060 Hannah Wang: for our clients.
197 00:19:11.060 ⇒ 00:19:15.080 Casie Aviles: It’s just gonna… you’re gonna specify there, like, which…
198 00:19:15.490 ⇒ 00:19:18.409 Casie Aviles: You know, elements are you going to…
199 00:19:18.410 ⇒ 00:19:19.060 Hannah Wang: I see.
200 00:19:19.060 ⇒ 00:19:24.420 Casie Aviles: on, for example, like, if you go on, like, inspect element, Yeah.
201 00:19:24.660 ⇒ 00:19:25.720 Casie Aviles: who knows?
202 00:19:25.860 ⇒ 00:19:30.790 Casie Aviles: Specify which selectors you want to interact with…
203 00:19:30.830 ⇒ 00:19:34.720 Hannah Wang: I see. Then it extracts the data, and then it writes that.
204 00:19:34.720 ⇒ 00:19:38.660 Casie Aviles: to Google Sheet, using, like, a… Connector, yeah.
205 00:19:39.110 ⇒ 00:19:39.860 Hannah Wang: Okay.
206 00:19:40.220 ⇒ 00:19:47.899 Hannah Wang: Gotcha. And then, polyatomic… Is… Is that similar to Daxter, or…
207 00:19:48.530 ⇒ 00:19:49.510 Casie Aviles: Yeah.
208 00:19:49.510 ⇒ 00:19:50.320 Hannah Wang: Okay.
209 00:19:50.320 ⇒ 00:19:59.259 Casie Aviles: Basically, Polytomic is… yeah, it’s similar to Dogster in the… but, like, it handles everything for us already. All we have to do is…
210 00:19:59.610 ⇒ 00:20:01.750 Casie Aviles: Like, set the configurations.
211 00:20:02.530 ⇒ 00:20:06.630 Casie Aviles: For Dagster, we had… we have to build, like, the pipeline.
212 00:20:07.300 ⇒ 00:20:11.810 Casie Aviles: The extraction pipeline from the source to the destination.
213 00:20:12.090 ⇒ 00:20:19.249 Casie Aviles: For Polyatomic, basically, it handles that for us. We’re just, you know, we just have to input, like.
214 00:20:20.340 ⇒ 00:20:23.559 Casie Aviles: The keys that we need, or, like, credentials.
215 00:20:23.970 ⇒ 00:20:26.990 Casie Aviles: So, yeah, Yeah.
216 00:20:29.060 ⇒ 00:20:29.730 Hannah Wang: Got it.
217 00:20:29.960 ⇒ 00:20:41.950 Hannah Wang: Okay, and then last thing, do you think there’s anything screenshotable that you… I can use in the…
218 00:20:42.930 ⇒ 00:20:46.920 Hannah Wang: The case study, or… Is it all just, like…
219 00:20:48.210 ⇒ 00:20:49.100 Casie Aviles: York.
220 00:20:49.100 ⇒ 00:20:50.230 Hannah Wang: Basically.
221 00:20:51.100 ⇒ 00:21:00.929 Casie Aviles: Yeah, it’s just mostly that. I think… yeah, that’s why I was trying to show this recording, I think this is… Yeah.
222 00:21:01.300 ⇒ 00:21:07.910 Casie Aviles: This one is the most… Presentable, I guess?
223 00:21:10.020 ⇒ 00:21:16.390 Casie Aviles: You can see, like, there’s the mouse pointer there, although there’s this pop-up that’s kind of blocking the view.
224 00:21:16.940 ⇒ 00:21:19.539 Casie Aviles: It wasn’t there when I was building it, but…
225 00:21:19.870 ⇒ 00:21:24.549 Casie Aviles: It basically, you know, you can see, like, there’s… there are things happening here.
226 00:21:25.560 ⇒ 00:21:28.470 Casie Aviles: That’s, that’s browser-based in action.
227 00:21:31.720 ⇒ 00:21:34.869 Casie Aviles: The rest, it’s just on… yeah, it’s just…
228 00:21:35.010 ⇒ 00:21:38.600 Casie Aviles: Mostly scheduling stuff, and just code, really.
229 00:21:38.700 ⇒ 00:21:40.670 Casie Aviles: And just a bunch of spreadsheets.
230 00:21:41.550 ⇒ 00:21:41.880 Casie Aviles: So.
231 00:21:41.880 ⇒ 00:21:42.660 Hannah Wang: Okay.
232 00:21:42.660 ⇒ 00:21:43.860 Casie Aviles: It’s not the most.
233 00:21:44.200 ⇒ 00:21:44.870 Casie Aviles: You know…
234 00:21:44.870 ⇒ 00:21:47.350 Hannah Wang: I don’t… Demoable, yeah.
235 00:21:47.350 ⇒ 00:21:47.790 Casie Aviles: Yeah.
236 00:21:47.790 ⇒ 00:21:49.380 Hannah Wang: That’s okay, no worries.
237 00:21:49.610 ⇒ 00:21:59.549 Hannah Wang: Okay, I think that is good. I’m going to… since this is… Luton wants this kind of urgently, I’m going to…
238 00:21:59.760 ⇒ 00:22:16.249 Hannah Wang: try to work on the case study, either today or tomorrow, and then I’ll ask you for feedback, and then, yeah, basically, what I did for the previous ones, and then we should be good to go, so… Okay. Okay, yeah, appreciate you, as always, Casey, and have a good day.
239 00:22:16.250 ⇒ 00:22:17.399 Casie Aviles: Alright, thank you, Hannah.
240 00:22:17.840 ⇒ 00:22:19.350 Hannah Wang: Bye.