Meeting Title: Zoom Meeting Date: 2025-05-09 Meeting participants: Annie Yu, Robert Tseng
WEBVTT
1 00:00:09.350 ⇒ 00:00:10.740 Annie Yu: Hello!
2 00:00:11.280 ⇒ 00:00:12.050 Robert Tseng: Hey! Annie!
3 00:00:13.570 ⇒ 00:00:19.280 Annie Yu: Well, Hi, well.
4 00:00:20.490 ⇒ 00:00:29.190 Annie Yu: yeah, wait. So so, okay, wait. So do I expect, like a weekly one on one with you and a wish, or I. I think I’m still confused. There.
5 00:00:29.190 ⇒ 00:00:49.143 Robert Tseng: Yeah, I mean, let’s let’s talk about like what you wanted to make work, I guess. Yeah, I I don’t know. Okay, well, I I know Oasius. He has expressed interest in wanting to become more of an engineering leader here, and so he’s like starting to meet up with people. I guess what he’s been doing doesn’t impact Hannah, and and
6 00:00:49.570 ⇒ 00:00:56.850 Robert Tseng: and amber or demote, because none of them report to him. It’s not like you report to him, either. We’re pretty flat structure. But like.
7 00:00:57.020 ⇒ 00:01:10.110 Robert Tseng: yeah, why don’t? We don’t need to have a redundant meeting either. But I’ve just been doing regular one on ones like weekly, one on ones with those folks and I guess we can make it, or we we don’t have to have one. We can.
8 00:01:10.490 ⇒ 00:01:19.579 Robert Tseng: I can just not. We could just not have it. Or if you want to make it something else, so that we’re not talking about the same thing multiple times. I’m totally okay with that as well.
9 00:01:20.760 ⇒ 00:01:21.710 Annie Yu: Yeah,
10 00:01:23.950 ⇒ 00:01:29.359 Annie Yu: sure, I think I’ll let you know. But I actually don’t have a meeting with a wish this week.
11 00:01:29.950 ⇒ 00:01:37.669 Annie Yu: so I I think we will be meeting later, but I don’t really know when will be the next one.
12 00:01:37.850 ⇒ 00:01:38.580 Robert Tseng: Okay.
13 00:01:39.020 ⇒ 00:01:39.660 Annie Yu: Yeah.
14 00:01:40.500 ⇒ 00:01:49.049 Robert Tseng: Okay? Well, yeah, I mean, I I guess. I was. I was like, kind of saying it kind of looks different for every person like.
15 00:01:49.250 ⇒ 00:01:58.539 Robert Tseng: for, like amber, she wants more coaching specifically on, like specific or like on on project management. I guess. So like
16 00:01:58.670 ⇒ 00:02:00.160 Robert Tseng: I feel like our
17 00:02:00.460 ⇒ 00:02:29.769 Robert Tseng: meetups are more like workshoppy for Hannah. We don’t talk about anything related to work. To be honest, it’s just what motivates her and thinks that she wants like, and she has some interest in some of the events that we’re planning, and so like. And then for Dame a lot, it’s it’s more like I’m connecting him to people, and he wants to grow in technical depth. And so, like, we’re doing like recommending books and like, we might read some stuff together or whatever. So I I kind of just share that as like a.
18 00:02:30.180 ⇒ 00:02:36.459 Robert Tseng: yeah, I mean, that’s my intention is so that you can have the space where we can kind of cover a wide range of topics. But
19 00:02:36.700 ⇒ 00:02:49.150 Robert Tseng: yeah, I mean, and if there’s like a way that you feel like you would like my time to be more focused and supporting you like I’m happy to kind of adjust it to. However, you feel like you would like to use this time.
20 00:02:49.830 ⇒ 00:02:54.450 Annie Yu: Okay. Okay? Then I think if that’s the case, I think I would
21 00:02:54.600 ⇒ 00:03:00.330 Annie Yu: love to maybe like explore and doesn’t have to be today. But like explore.
22 00:03:01.140 ⇒ 00:03:12.510 Annie Yu: like, with our current clients or data like how we could turn something into like a project that I can like apply a machine learning that kind of thing, so it could be like
23 00:03:12.830 ⇒ 00:03:18.370 Annie Yu: supervised learning like predict whether a customer will repurchase or unsubscribe, or like.
24 00:03:18.370 ⇒ 00:03:19.220 Robert Tseng: Yeah, yeah.
25 00:03:19.220 ⇒ 00:03:25.639 Annie Yu: Alice is like finding items that are often bought together, so that that’ll be fun.
26 00:03:26.010 ⇒ 00:03:53.950 Robert Tseng: Okay, yeah. I know, we kind of like had some stuff teed up like, and it never really made it into the cycle. For the Ed mentions. Specifically like, predicted Ltv, and like, and we kind of also thought about like doing incrementality measurement. So yeah, I mean, if it’s helpful to just like, take time regularly to just go and talk about these Ml applications to for like specific clients that you’re working on. Yeah, I’m happy to do that, because I think
27 00:03:54.250 ⇒ 00:04:11.139 Robert Tseng: those things are like less, always less urgent, but like obviously super high value. And I don’t feel like I get enough time to talk about them during stand ups. So I mean, yeah, if that’s if that’s where we can spend our time. Like I I’d I’d be happy to do that.
28 00:04:12.020 ⇒ 00:04:25.870 Annie Yu: Sounds good. But personally, also, just like I’m not super interested in marketing data compared to like like pos sales, data, or or, you know, like supply chain, that kind of thing. But I think that’s
29 00:04:26.200 ⇒ 00:04:30.559 Annie Yu: also like it wouldn’t be a deal breaker to deal with marketing data. So.
30 00:04:31.620 ⇒ 00:04:42.900 Robert Tseng: Yeah, I mean, with the marketing data stuff, I mean, incrementality measurement, like, sure, we don’t have to touch that like, if you would prefer to just do now that they even just launch like 13 products, like.
31 00:04:43.740 ⇒ 00:04:49.799 Robert Tseng: they’re yeah, they’re gonna care more about like product bundling. And yeah, I mean.
32 00:04:51.140 ⇒ 00:04:58.930 Robert Tseng: They want to know who the highest like who’s likely to repurchase and like stuff like that, like I I think that’s that’s important.
33 00:04:59.350 ⇒ 00:05:17.649 Robert Tseng: I can loop you into those conversations if you want to be more. But that is kind of marketing like it’s the lifecycle marketing team is trying to retarget existing customers or target people on their email list to try to upsell them into certain things. So like, I feel like anything point of sale related does like get.
34 00:05:18.130 ⇒ 00:05:26.220 Annie Yu: No, I think that makes sense like to identify like who’s likely to to turn next month, and then we can set them personalized.
35 00:05:26.500 ⇒ 00:05:27.160 Robert Tseng: Yeah.
36 00:05:27.160 ⇒ 00:05:30.120 Annie Yu: And that that totally makes sense.
37 00:05:31.560 ⇒ 00:05:45.669 Annie Yu: Sure. I think I yeah, I think I do wanna like explore areas. Like that. But also like acknowledging that it’s gonna be like. So these kind of projects will be like long, longer term. And like.
38 00:05:45.670 ⇒ 00:05:46.230 Robert Tseng: Yeah.
39 00:05:46.230 ⇒ 00:05:50.790 Annie Yu: With like, not like super quick turnaround. So so, yeah.
40 00:05:51.500 ⇒ 00:06:02.500 Robert Tseng: Yeah, yeah, no. I mean, I I think everything will take at least 2 weeks to like kind of stuff. If you run and run an experiment. It’s gonna take like 2 weeks to get enough data that you to see something significant. So,
41 00:06:03.260 ⇒ 00:06:04.530 Robert Tseng: yeah, like.
42 00:06:05.770 ⇒ 00:06:13.449 Robert Tseng: okay, I will keep that in mind. I I think the so. One thing I’ll mention, so I I guess maybe you’ve heard some
43 00:06:13.550 ⇒ 00:06:18.660 Robert Tseng: tooling decision. So with segment I guess
44 00:06:18.780 ⇒ 00:06:22.490 Robert Tseng: I’m trying to work with the lifecycle marketer now on Eden. And
45 00:06:22.730 ⇒ 00:06:37.900 Robert Tseng: basically I’ve looked. I’ve studied like his campaigns and what he runs. We can look at it together and make record a loom for you, and I’ll give you. You can log into customer, I/O, and stuff so you can kind of you should. Probably I think it’d be good to familiarize yourself with, like how he makes campaigns
46 00:06:38.576 ⇒ 00:06:46.989 Robert Tseng: things that he’s using. He just looks at like, most recent, like, everything is like time. It’s like time based triggers. So it’s like.
47 00:06:47.447 ⇒ 00:06:53.340 Robert Tseng: send them an E send them an email. Send an email to somebody who’s who’s ordered 3 plus
48 00:06:53.450 ⇒ 00:06:59.938 Robert Tseng: like times in the past 30 days, like, there’s nothing predictive about it. Everything is just like,
49 00:07:00.750 ⇒ 00:07:06.652 Robert Tseng: yeah, it’s it’s just like, based on recency and frequency of like their of like purchases, or whatever.
50 00:07:07.160 ⇒ 00:07:11.250 Annie Yu: Like a Cdp to to trigger those.
51 00:07:11.840 ⇒ 00:07:30.230 Robert Tseng: Yeah, well, kind of so so segment is the Cdp right now we’re like, I’m evaluating whether or not to keep it. My, I’m leaning towards no because he uses customer. I/OI don’t know if you’ve used other ceps like I mean, maybe salesforce at Microsoft or right this Microsoft use salesforce. I forgot what they use.
52 00:07:30.480 ⇒ 00:07:31.430 Robert Tseng: I don’t.
53 00:07:31.430 ⇒ 00:07:33.846 Annie Yu: So I don’t even know.
54 00:07:34.330 ⇒ 00:07:43.299 Robert Tseng: Okay? Well, anyway, like a customer engagement platform like a Mailchimp raise customer, I/O is there just like email and SMS platforms that like
55 00:07:43.470 ⇒ 00:07:45.279 Robert Tseng: they they get.
56 00:07:45.610 ⇒ 00:07:50.790 Robert Tseng: Yeah, you, you push some sales data and like some customer data into there so that
57 00:07:50.890 ⇒ 00:08:12.165 Robert Tseng: you can, you know, make these filters based on like number of orders. Type of like order, like product name stuff like that. But there’s nothing. There’s no predictive piece to it. There’s no like filter by customers that are have like high, intense, or like high likelihood of being in your top. Ltv. Bar, or whatever
58 00:08:12.520 ⇒ 00:08:25.295 Robert Tseng: So I think the goal, then, would be to like, create some of those features so that we can push it into these. Cp be like alright. This is the cohort of customers that’s likely to be in your highest Ltv bucket.
59 00:08:25.900 ⇒ 00:08:32.990 Robert Tseng: and like that could be an example of like how we bring a predictive element to the campaign. And so
60 00:08:33.659 ⇒ 00:08:41.130 Robert Tseng: yeah, we kind of have to think about like, what are those other cuts? And we are kind of, I’m like evaluating that right now with the lifecycle person.
61 00:08:42.131 ⇒ 00:08:52.570 Robert Tseng: But yeah, I think that would probably be like where this Ml thing would go. So I might start looping you into those emails and conversations.
62 00:08:54.120 ⇒ 00:08:55.840 Robert Tseng: Yeah, so.
63 00:08:56.640 ⇒ 00:09:06.350 Robert Tseng: But I I think that that would be that that would be a really good project to work on like nobody. Nobody is touching that right now. We don’t have to do incrementality. First, st because
64 00:09:06.480 ⇒ 00:09:15.589 Robert Tseng: we did buy an incrementality tool, and whether it works or not like they’re gonna just use it for now, and like I’d rather us do something that new.
65 00:09:16.540 ⇒ 00:09:22.690 Annie Yu: Yeah, sounds good. And I can always start by like, just learning the data. So, okay.
66 00:09:22.700 ⇒ 00:09:25.809 Robert Tseng: Okay, cool. Yeah. Then I will make that happen.
67 00:09:26.210 ⇒ 00:09:40.579 Annie Yu: And one thing I just wanna like voice is about real data. I think I think I think I I don’t. I I did add you when they shared the update.
68 00:09:40.580 ⇒ 00:09:43.110 Robert Tseng: Yeah, I didn’t read too closely. What was that about?
69 00:09:43.110 ⇒ 00:09:45.580 Robert Tseng: They don’t have horizontal bar charts. Yeah.
70 00:09:45.580 ⇒ 00:10:10.610 Annie Yu: Yeah, and that’s just one thing. But I think I don’t like how their answer is like we could add this, but we will enable you, we will let you like. Do your full customization enable you to do like Vega Lite, which I think Mega Lite is, is all right, but I I think the fact that they are saying like, you can use AI and generate code, copy and paste. And I I just don’t like.
71 00:10:10.940 ⇒ 00:10:12.899 Annie Yu: I think I think for me.
72 00:10:13.570 ⇒ 00:10:17.660 Annie Yu: then I like. With that I just failed
73 00:10:18.140 ⇒ 00:10:38.080 Annie Yu: to see the value in like choosing real over other more mature and powerful bi tools, and like with, if I want to write a code, if I want to write code to build my visualizations, I can use python. There’s like Vega Altair, which is basically like Vega lighting python. And like
74 00:10:38.530 ⇒ 00:10:48.400 Annie Yu: Seabourne, is also very like respected. I think I just don’t I? And I’m like being open-minded, but I just don’t see the value right now. And yeah, and
75 00:10:48.500 ⇒ 00:11:08.379 Annie Yu: and I think honestly, it’s like slowing me down more than it’s helping compared to other tools that I’m using. But I I also like, believe it may have its place. Maybe it’s easier to like a data engineer to transition. But I don’t know as like a non data engineer. I think it’s like slowing me down more than it’s helping. But.
76 00:11:08.800 ⇒ 00:11:18.980 Robert Tseng: I agree. I I’m I also. I agree with you. I think. So I think. Let me just like kind of share some context on like, why real? So like.
77 00:11:19.210 ⇒ 00:11:29.739 Robert Tseng: yeah, I think, yeah, like the whole bi as code, like
78 00:11:29.930 ⇒ 00:11:39.440 Robert Tseng: movement with real and like, like, dash being like, part of that is definitely more to help data engineers like do data visualization.
79 00:11:39.910 ⇒ 00:11:41.930 Robert Tseng: right? Because before, like.
80 00:11:43.030 ⇒ 00:11:48.126 Robert Tseng: I mean, yeah, like it to them, it’s it’s it’s like it’s, it’s helped kind of create that
81 00:11:48.590 ⇒ 00:12:04.639 Robert Tseng: that ability. But it doesn’t help like data scientists or data analysts who are already using being used to being downstream up like make make more stuff. So like, I think that’s I don’t think you’re the ideal user of of real. To be honest, like, I think it’s
82 00:12:04.870 ⇒ 00:12:19.870 Robert Tseng: it’s easy for our team to set up, because, like, it just is just like automatically generated on top of the Dbt models. And it’s like an easy 1st layer of like just adding, like pivot table functionality to all of your underlying data models.
83 00:12:20.754 ⇒ 00:12:32.960 Robert Tseng: And so that’s why we stand it up for every client. I’m assuming that you’re talking about. For, like ABC, it’s like any of these clients where we don’t want to make that much investment in complex visualization. We’re not going to bring tableau to ABC, right?
84 00:12:33.252 ⇒ 00:12:51.059 Robert Tseng: Or at least not right now. And yeah, like. So that’s what reals there, for it’s like to let them like view some trends. And like, do some do some cuts of their data themselves. But yeah, if you prefer to just like export data. And like, do your analysis in python or elsewhere, like, don’t feel like you’re constrained to real
85 00:12:51.488 ⇒ 00:12:58.309 Robert Tseng: like with pool parts. Right now, like, I basically like Amber has the same frustration as well. So
86 00:12:58.540 ⇒ 00:13:04.580 Robert Tseng: I just told her to export the data and just do it. She just doesn’t in excel. It’s like totally fine. I don’t care, so do.
87 00:13:04.580 ⇒ 00:13:18.680 Annie Yu: The thing is, I have to load snowflakes table into real, and then did my like kind of modeling there. And even if I want to explore that model. I have to do that within that, and just the whole thing.
88 00:13:19.590 ⇒ 00:13:21.630 Robert Tseng: Okay, yeah. I mean, why, like.
89 00:13:22.270 ⇒ 00:13:25.290 Robert Tseng: is there like, why, why do you have to do it in real like, why don’t you just
90 00:13:25.970 ⇒ 00:13:36.039 Robert Tseng: why don’t you just like export the model to and just why don’t you just take X snowflake data and you can run python on top of snowflake. It’s like, why, why go through real.
91 00:13:36.870 ⇒ 00:13:41.719 Annie Yu: I I don’t know. I think that’s just how it is. And like.
92 00:13:42.120 ⇒ 00:13:49.850 Annie Yu: so every edit on real could be like reviewed as a pr, I think that’s also one thing.
93 00:13:51.270 ⇒ 00:13:58.869 Robert Tseng: Okay. But the stuff that you’re building like it, it becomes like a dashboard or something that they’re that they’re that they’re using regularly or like.
94 00:13:59.120 ⇒ 00:14:14.660 Robert Tseng: why are you doing like new analysis like to me? Real is like after there’s like a clear like view that you tried to set up. And you. Then you spend it up like I’m not really following like, why, you’re use having to use real in your analysis like that as your analysis tool.
95 00:14:19.430 ⇒ 00:14:22.836 Robert Tseng: So I mean, maybe we should make this bit more like tangible. Let me
96 00:14:23.530 ⇒ 00:14:27.800 Robert Tseng: I can run. I can run a slightly over. Let me give me an example.
97 00:14:27.800 ⇒ 00:14:31.920 Annie Yu: I’m gonna need to head out later.
98 00:14:31.920 ⇒ 00:14:32.809 Robert Tseng: That out. Okay.
99 00:14:33.080 ⇒ 00:14:38.579 Robert Tseng: alright, let me just yeah, quickly here. So this is the pull parts real. Right? So like, there’s a few different things that are set up
100 00:14:38.750 ⇒ 00:14:42.590 Robert Tseng: like, yeah, like, this is like their paid marketing person
101 00:14:43.270 ⇒ 00:14:58.469 Robert Tseng: like, yeah, she she, this is their campaign manager, like they built them a dashboard in real. She just uses this. That’s fine. She never asked for anything new that this works for her like. She can look at her channel like performance, and she can look at her campaign for performance. It’s good
102 00:14:59.086 ⇒ 00:15:24.020 Robert Tseng: we’ve set up some other like view set up for them daily. Kpis. This is what the executives look at. They get to look at like daily daily sales and everything. So like my point is like, once there’s like a clear like view of the data that we want. Then then we set it up in real, but like to get there, we don’t need to be like experimenting and and real rightly like, I know, like, this is probably just
103 00:15:24.370 ⇒ 00:15:31.559 Robert Tseng: like there’s a couple of models and Dbt, I would just like, you know, take cuts of like small cuts of that, and then I would export it.
104 00:15:32.030 ⇒ 00:15:38.769 Robert Tseng: Try to do the analysis in sequel, kind of like what I did with with Eden, and I just sent you a query like, I was just like trying to
105 00:15:39.100 ⇒ 00:15:50.419 Robert Tseng: make some stuff like. Figure out what the what the table was like I, I would just do like sequel or python based analysis and not even bother with real until, like there’s a clear like view that I’m building towards.
106 00:15:52.560 ⇒ 00:15:53.059 Annie Yu: Did I make.
107 00:15:53.060 ⇒ 00:15:59.750 Robert Tseng: I don’t know if that would speed it up for you, but that’s well. From what you’re describing to me, it does sound like it’s kind of slow if you have to do it that way.
108 00:16:00.320 ⇒ 00:16:03.200 Annie Yu: Yeah, yeah, okay.
109 00:16:03.200 ⇒ 00:16:09.339 Robert Tseng: So maybe there’s just like a like a save real for the end, like the last mile piece, like.
110 00:16:09.620 ⇒ 00:16:11.080 Robert Tseng: Yeah, I understand that you can.
111 00:16:11.080 ⇒ 00:16:11.660 Annie Yu: I also.
112 00:16:11.660 ⇒ 00:16:12.070 Robert Tseng: To change.
113 00:16:12.070 ⇒ 00:16:21.409 Annie Yu: I join within real, too. So we are joining 2 tables from Snowflake, and then make it into a model unreal.
114 00:16:22.400 ⇒ 00:16:36.980 Robert Tseng: Yeah, I’m saying like, why don’t you just do that in Snowflake? Do the sequel worksheet. Just build it out. And then, like, if you need to. Once you have the table. If you don’t feel comfortable visualizing in sequel, and just like export it, use a python notebook, or whatever like. Get the thing that you want.
115 00:16:37.300 ⇒ 00:16:42.570 Robert Tseng: And like, yeah, if that’s a faster workflow for you to just get something in front of the client.
116 00:16:43.152 ⇒ 00:16:49.040 Robert Tseng: Then we can afterwards, like you’ll know exactly what needs to be built and real and like
117 00:16:49.280 ⇒ 00:16:57.219 Robert Tseng: it doesn’t even have to be you. It could be like the the analytics engineering team that goes and like builds that like view in real.
118 00:16:57.850 ⇒ 00:17:01.710 Robert Tseng: because it’s all like in like Bi as code. Anyway, right?
119 00:17:03.210 ⇒ 00:17:05.790 Annie Yu: yeah, that’s a good idea.
120 00:17:06.040 ⇒ 00:17:19.620 Robert Tseng: Yeah. So I’m just saying, like, don’t limit yourself to using real for every piece of analysis. Like, I understand, there’s like different stages to analysis for definitely like joins. And like combining data, I feel like sequel is always the fastest, like, you might as well just do SQL
121 00:17:20.048 ⇒ 00:17:24.499 Robert Tseng: and then for visualization, like, if you’re faster in python, just just use python.
122 00:17:25.290 ⇒ 00:17:37.420 Annie Yu: Yeah. But I think one thing, though, with, if I do my like, join in Snowflake, like there’s no way like, how do I make people to review that, or
123 00:17:38.330 ⇒ 00:17:47.250 Annie Yu: if I have to make a change, because right now I do have to go through Pr, and then we’ll review it. But if I remove that everything
124 00:17:47.650 ⇒ 00:17:52.940 Annie Yu: to Snowflake, then there’s no like a review stage that’s forced.
125 00:17:54.370 ⇒ 00:18:00.536 Robert Tseng: Yeah, I mean, that’s like, once, you actually make the change in real. So yeah, I I see what you’re saying.
126 00:18:02.290 ⇒ 00:18:11.549 Robert Tseng: I mean, this ends up being like a devops kind of like problem. I I don’t know. And you’re not like changing Dbt models. I mean, I’ve seen you push prs like on Eden side, or whatever like.
127 00:18:11.790 ⇒ 00:18:17.800 Annie Yu: Yeah, but I don’t think a ABC. Wait. I don’t think ABC is going through Dbt.
128 00:18:18.510 ⇒ 00:18:21.760 Annie Yu: so we’re using real as a Dbt substitute.
129 00:18:22.520 ⇒ 00:18:23.420 Robert Tseng: I see.
130 00:18:26.340 ⇒ 00:18:29.009 Robert Tseng: I mean, I can’t imagine. These are like, super.
131 00:18:29.320 ⇒ 00:18:33.979 Robert Tseng: It’s here. Okay? Well, I will. You know what I know. You gotta go, so I’ll
132 00:18:34.100 ⇒ 00:18:42.040 Robert Tseng: I’ll try to like. I’ll I’ll talk to you, Tom. I’ll see if like. Can you just like, can you just do reviews just like in tickets like, kind of like what we didn’t eaten like.
133 00:18:42.410 ⇒ 00:18:53.069 Robert Tseng: I just send you the query, you know. And like I, I commented the code like, I tell you what I what I need you to look at. And you just we’re just. It’s it’s not exactly like our views or whatever, but
134 00:18:53.500 ⇒ 00:18:57.600 Robert Tseng: like that, that should be okay until you until you get to the final.
135 00:18:57.780 ⇒ 00:18:59.739 Robert Tseng: Until you get to the final Pr, I don’t know.
136 00:19:00.520 ⇒ 00:19:22.340 Annie Yu: Yeah, no, that makes sense. Yeah, I’ll see. I yeah, I’m not saying like, we, we can’t use it. I just think. And I think with what at least what ABC has now, it’s not a ton. So I think it’s fine. But if there are gonna like skill, or or whatever like in the future. I just I I can’t. I can use that in the long term like for, like building more dashboards.
137 00:19:23.570 ⇒ 00:19:25.063 Robert Tseng: Okay, yeah.
138 00:19:26.230 ⇒ 00:19:30.120 Robert Tseng: okay, yeah. I’ll I’ll I’ll talk to Tom about it. Let me let me look into it. Yeah.
139 00:19:30.400 ⇒ 00:19:36.379 Annie Yu: Yeah, but it’s not super serious. I will still work with it. Okay.
140 00:19:36.380 ⇒ 00:19:36.740 Robert Tseng: Yeah.
141 00:19:38.330 ⇒ 00:19:38.949 Robert Tseng: All right.
142 00:19:38.950 ⇒ 00:19:43.890 Annie Yu: I’m gonna I’m gonna head out grabbing a coffee so.
143 00:19:43.890 ⇒ 00:19:44.430 Robert Tseng: Okay.
144 00:19:44.750 ⇒ 00:19:45.150 Annie Yu: I’ll see.
145 00:19:45.150 ⇒ 00:19:45.810 Robert Tseng: Bye.
146 00:19:45.810 ⇒ 00:19:47.269 Annie Yu: Talk to. You soon. Have a good one.
147 00:19:47.930 ⇒ 00:19:48.899 Robert Tseng: Yeah, you too.