Meeting Title: Friday Brainforge Demos & Retro Date: 2025-12-01 Meeting participants: Awaish Kumar, Henry Zhao, Rico Rejoso, Amber Lin, Casie Aviles, Mustafa Raja, Demilade Agboola, Ryan Brosas, Hannah Wang, Holly Condos, Robert Tseng


WEBVTT

1 00:02:48.900 00:02:51.479 Henry Zhao: Hi, Awish, how are you doing? Where’s everybody?

2 00:02:56.880 00:02:59.709 Awaish Kumar: Hello, I’m good, how are you?

3 00:03:00.330 00:03:01.170 Henry Zhao: Good, thanks.

4 00:03:51.680 00:03:52.870 Amber Lin: Hi there.

5 00:03:53.770 00:03:55.950 Amber Lin: Is this meeting still happening?

6 00:03:55.950 00:03:58.850 Henry Zhao: Yes, and I’m looking forward to our slide session.

7 00:04:04.470 00:04:09.910 Amber Lin: Awesome. Do we know if Utum’s back from ABC? Because he’s still on-site.

8 00:04:10.860 00:04:11.720 Henry Zhao: Oh, really?

9 00:04:11.720 00:04:16.630 Amber Lin: Earlier, I know he was on the site, but I don’t know where he’s… where he is now.

10 00:29:08.840 00:29:10.030 Amber Lin: Hi, Robert.

11 00:29:10.210 00:29:11.939 Amber Lin: Do you know if it has come into that?

12 00:29:11.940 00:29:12.940 Robert Tseng: What’s going on here?

13 00:29:13.870 00:29:14.430 Amber Lin: Oh, he’s…

14 00:29:14.430 00:29:15.330 Robert Tseng: still on site.

15 00:29:15.840 00:29:18.470 Amber Lin: Oh, that’s so funny. Okay.

16 00:29:18.730 00:29:24.530 Amber Lin: Well, we started… we didn’t start, because we weren’t sure if he would be coming.

17 00:29:24.670 00:29:29.330 Amber Lin: But we still have some time left. Is there any…

18 00:29:29.450 00:29:35.459 Amber Lin: Key, like, exec stuff we want to go over, any new sales that’s happening we want to announce?

19 00:29:36.320 00:29:42.620 Robert Tseng: I… did not prep for this, but I guess,

20 00:29:43.280 00:29:53.120 Robert Tseng: Yeah, I mean, we have a few things kicking off. Remo closed, so that’s… That’s where we’re,

21 00:29:54.060 00:29:59.700 Robert Tseng: That’s… That’s starting today. And then,

22 00:29:59.950 00:30:09.060 Robert Tseng: Lilo Socials, Lilo Socials started today, Element starting this week, and

23 00:30:10.130 00:30:26.439 Robert Tseng: I guess, like, he rose up for renewal on Wednesday, so… and then Utam’s obviously at ABC, kind of doing discovery there. So, yeah, I mean, we just pretty much have, like, 5 new clients, like, all starting this week, so we’re kind of just swamped on delivery side.

24 00:30:28.590 00:30:34.520 Robert Tseng: Yeah, I guess that’s… that’s a mix of the… I mean, and then as far as, like, new…

25 00:30:34.940 00:30:41.499 Robert Tseng: new deals, I mean, we are… we are continuing to talk to new,

26 00:30:41.640 00:30:55.709 Robert Tseng: to talk… talk to leads. I mean, nothing… nothing noteworthy right now. I think these are all stations for… for next year, in… in Q1 2026. But yeah, I guess that’s… that’s kind of what we’ve been working on.

27 00:30:58.830 00:30:59.485 Amber Lin: Mmm…

28 00:31:12.690 00:31:22.200 Amber Lin: I guess, is there specific demos? Any teams that, I guess specifically the AI team, is there anything we want to demo, and…

29 00:31:22.350 00:31:23.180 Amber Lin: Show.

30 00:31:30.960 00:31:33.219 Casie Aviles: Didn’t really prep for any demos.

31 00:31:34.510 00:31:35.060 Amber Lin: I see.

32 00:31:35.060 00:31:40.020 Casie Aviles: for… Yeah, but we’ve just been working on…

33 00:31:40.860 00:31:52.890 Casie Aviles: Improvements on the platform, specifically the ticket generation, like, after the meetings, and… We’re also, like… planning… on…

34 00:31:53.310 00:31:56.350 Casie Aviles: Doing some migration internally for, like.

35 00:31:58.060 00:32:06.379 Casie Aviles: You know, for our AI agents, so our client hub agents, so we’ll be moving them to code, so before, we were just using

36 00:32:07.370 00:32:12.039 Casie Aviles: And even… So yeah, hopefully, like, this, this will…

37 00:32:12.870 00:32:21.630 Casie Aviles: allow us to, like… and also we’re kind of doing the same thing for ABC, in order to… since anytime we’re running into some limits already, so…

38 00:32:22.140 00:32:30.389 Casie Aviles: yeah, moving it into code should ideally help us, you know, address some observability issues with the AI.

39 00:32:31.180 00:32:33.399 Casie Aviles: Be able to do triage.

40 00:32:33.920 00:32:35.849 Casie Aviles: Faster, so yeah.

41 00:32:36.060 00:32:37.980 Casie Aviles: That’s pretty much it that we have.

42 00:32:40.290 00:32:43.020 Amber Lin: Cool. Guess…

43 00:32:43.540 00:32:53.100 Amber Lin: Then, on the marketing side, is there any new case studies that we’re putting out? In the last week on the delivery meeting, we outlined a few case studies.

44 00:32:53.200 00:32:58.719 Amber Lin: Do we know who to ask for those case studies?

45 00:32:58.870 00:33:02.799 Amber Lin: it’s like, what case studies are lined up.

46 00:33:02.950 00:33:05.420 Amber Lin: I feel like everybody has a few things that…

47 00:33:05.590 00:33:10.139 Amber Lin: can be made into case studies, I just want to know if they’re happening yet.

48 00:33:11.330 00:33:14.150 Hannah Wang: You can…

49 00:33:14.480 00:33:22.639 Hannah Wang: probably message me, and then I would just need to get it prioritized, because I don’t really know which one is…

50 00:33:22.880 00:33:27.270 Hannah Wang: More urgent, or would be the most useful.

51 00:33:27.720 00:33:37.509 Hannah Wang: And then I think with the AI team’s new case study assistant, I wouldn’t need to interview people anymore, it’s just people record their,

52 00:33:38.270 00:33:42.319 Hannah Wang: Just them talking with the assistant, so that should be pretty…

53 00:33:42.480 00:33:46.820 Hannah Wang: Easy. I know you messaged me, a number of case studies.

54 00:33:46.920 00:33:54.550 Hannah Wang: Amber, and then… yeah, there’s, like, a backlog of case studies we have in, linear.

55 00:33:54.710 00:33:57.390 Hannah Wang: But… Yeah, I guess…

56 00:33:57.650 00:34:08.929 Hannah Wang: I would just need help prioritizing it. And I don’t really know who else to ask except Utam and Robert, but if… in terms of priority,

57 00:34:09.830 00:34:14.880 Hannah Wang: So yeah, like, I can… I can have the list, it’s just I don’t know which ones to work on first.

58 00:34:16.000 00:34:19.409 Amber Lin: Cool. Okay.

59 00:34:19.670 00:34:25.409 Amber Lin: I know we talked about it, like, a week ago, so hopefully people know…

60 00:34:25.590 00:34:27.949 Amber Lin: What case studies they have on hand.

61 00:34:28.270 00:34:30.130 Amber Lin: Okay.

62 00:34:32.120 00:34:33.199 Amber Lin: I did…

63 00:34:33.469 00:34:43.919 Amber Lin: Sorry, I did prepare slides, and I know we were just sitting in this room, and I went to do other analysis. I can go over… I have something to present, I have a…

64 00:34:44.070 00:34:54.019 Amber Lin: AI-assisted Analysis Outline. Are people interested in seeing that? I can show you what I’ve been doing and how I make the slides.

65 00:34:55.830 00:34:57.389 Hannah Wang: Yeah, let’s do it.

66 00:34:57.930 00:34:58.560 Amber Lin: Okay.

67 00:35:02.270 00:35:03.060 Amber Lin: Cool.

68 00:35:10.590 00:35:16.260 Amber Lin: So my simple process… is…

69 00:35:16.560 00:35:20.049 Amber Lin: I go from the database to cursor.

70 00:35:20.250 00:35:23.939 Amber Lin: And then from cursor, I build the narrative.

71 00:35:24.130 00:35:32.180 Amber Lin: Then the third, like, the third round, I make some recommendations, and finally put it into slides. And when I…

72 00:35:32.940 00:35:39.350 Amber Lin: Let’s see, okay. So, starting off with cursor… I usually…

73 00:35:39.580 00:35:53.440 Amber Lin: this is the database that we see. I think there’s a lot of tables, there’s a lot of stuff going on, and so it’s… how do we start from there to give out something that

74 00:35:53.550 00:36:00.040 Amber Lin: Clients would read that has… has meaning, or has some insights that’s interesting.

75 00:36:00.840 00:36:09.670 Amber Lin: And so, starting from a database. I connect it. I connect to it via cursor.

76 00:36:09.760 00:36:25.580 Amber Lin: And this step, I think you can either ask our more technical folks for assistance to set up cursor, I needed help there. And then you can search online to connect to DuckDB or connect to Snowflake, depending on

77 00:36:25.910 00:36:28.120 Amber Lin: Where the class database is.

78 00:36:29.060 00:36:48.780 Amber Lin: And then, the second step would be to explore the database, and here’s something really simple you can do it just to tell ChatGPT, tell Cursor, can you explore this database? And then you can find the tables, and you can say, can you explore specific tables?

79 00:36:49.440 00:36:51.280 Amber Lin: And then it will give you

80 00:36:51.530 00:36:58.049 Amber Lin: The different fields, the dimensions, if there’s nulls. So, you can give it a very generic

81 00:36:58.760 00:37:07.499 Amber Lin: Argument, and then something… you might… you will be able to have a broad view of what data you actually have.

82 00:37:08.370 00:37:17.580 Amber Lin: And I think that’s the two preparation steps, and the next part does it get a bit more interesting, which in this part is where we look at

83 00:37:17.720 00:37:37.650 Amber Lin: one dimension of, maybe it’s sales, maybe it’s traffic, and we look at, is there something interesting, or to say if there’s something really big or really small. I think this is for people who have not done analysis before, if, essentially.

84 00:37:37.740 00:37:45.419 Amber Lin: The first step is to see, okay, is there… abnormal patterns. Sometimes.

85 00:37:46.760 00:37:50.689 Amber Lin: Sometimes really big things, like here.

86 00:37:50.940 00:38:05.620 Amber Lin: like, the green part got really big, that’s something interesting. Or if it suddenly dipped really small, that’s interesting. Or even if it’s flat over time, you can ask, okay, why is it…

87 00:38:05.620 00:38:13.620 Amber Lin: flat over time. Say, if a company’s revenue is flat over 3 years, then that’s something we can point out.

88 00:38:14.080 00:38:23.990 Amber Lin: Alright, so in this step, essentially, it’s to see, are there patterns noteworthy patterns? Too big? Too small?

89 00:38:24.200 00:38:26.799 Amber Lin: Or… no change at all.

90 00:38:27.480 00:38:36.259 Amber Lin: So I think that’s a pretty basic rule of thumb you can apply there. You can look at it from one dimension, and then you can compare

91 00:38:36.510 00:38:38.889 Amber Lin: multiple dimensions.

92 00:38:39.150 00:38:41.199 Amber Lin: And put it on the same graph.

93 00:38:41.410 00:38:56.179 Amber Lin: over… over time, you can… and then, to spice it up, you can slice two metrics. You can say, okay, let’s look at product category, and then look at it over time.

94 00:38:56.660 00:39:00.230 Amber Lin: So, I think those are some basic

95 00:39:00.620 00:39:11.889 Amber Lin: approaches to look at the data to see if there’s any patterns. I think that’s a good place to start. And then, I think, once you complete this step with Cursor, you would end up with

96 00:39:12.320 00:39:19.510 Amber Lin: Some graphs, some insights, some numbers, and at that point, it’s,

97 00:39:19.870 00:39:26.029 Amber Lin: how can we make sense of what’s happening? Because if you only have a graph.

98 00:39:26.560 00:39:31.240 Amber Lin: What does that… what does that mean? What does it mean for…

99 00:39:32.150 00:39:37.759 Amber Lin: the client, what can they do with it? Like, why should they… why should they care about

100 00:39:38.650 00:39:41.570 Amber Lin: what you just found, and I think that’s the…

101 00:39:41.900 00:39:51.549 Amber Lin: That’s the key step that I think a lot of us struggle with, is what do we make of… what do we make of the image, or make of the data that we have?

102 00:39:52.600 00:39:53.670 Amber Lin: And…

103 00:39:53.930 00:40:01.130 Amber Lin: From there, I think once we have the data, there’s 3 steps we need to do. So, first is

104 00:40:01.310 00:40:10.710 Amber Lin: What story does it tell? And then, based on that, what recommendations can we do?

105 00:40:10.750 00:40:23.250 Amber Lin: That makes it actionable, and that makes it… makes it so that the client can do something about it. And lastly, we can put it together, because often in this step.

106 00:40:23.250 00:40:36.290 Amber Lin: especially when we use cursor, I think there’s a lot of information that will come out, but our brain can only digest so much, so our job here is to filter it down so that the client can only see

107 00:40:36.370 00:40:39.659 Amber Lin: Like, 3 relevant points instead of 30.

108 00:40:40.690 00:40:48.129 Amber Lin: So… I think here… so the first step is to figure out if we can tell a story.

109 00:40:48.240 00:40:56.030 Amber Lin: And Like, if we just look at this here… What do you think…

110 00:40:56.160 00:40:58.170 Amber Lin: Like, what do you think this…

111 00:40:58.650 00:41:02.619 Amber Lin: this would mean, right? What trends do you see, and…

112 00:41:02.880 00:41:05.210 Amber Lin: what… what might it mean? Can…

113 00:41:05.450 00:41:10.699 Amber Lin: Can I grab someone to tell me about that? I’m gonna point to a participant.

114 00:41:11.330 00:41:14.730 Amber Lin: Can someone volunteer, or I will call names.

115 00:41:19.670 00:41:25.260 Amber Lin: Okay, I will call names. Let’s see…

116 00:41:25.680 00:41:28.860 Amber Lin: Will Mustafa help me here?

117 00:41:32.800 00:41:36.989 Mustafa Raja: Sorry to inconveni, ask the question again?

118 00:41:37.710 00:41:43.299 Amber Lin: What do you… is there a trend that you see from this, and what do you think it would mean?

119 00:41:44.770 00:41:45.910 Mustafa Raja: Let me see…

120 00:42:03.170 00:42:05.540 Mustafa Raja: Hmm… I can’t see it.

121 00:42:27.040 00:42:36.119 Amber Lin: I think the conclusion that I drew back then was that, we saw this blue trend going up, so there was…

122 00:42:36.290 00:42:39.970 Amber Lin: It started pretty small, and then it grew pretty big.

123 00:42:40.050 00:42:56.729 Amber Lin: And then we could say that, okay, as people started to make their second, third, fourth, like, they… as they made more purchases, as they returned to make more purchases, they started… more of them started getting classic cookies.

124 00:42:57.060 00:42:57.950 Mustafa Raja: Instead of before.

125 00:42:57.950 00:43:05.859 Amber Lin: Or when, usually for people’s first order, they generally only, say, like, 20% of them get classic cookies.

126 00:43:05.990 00:43:08.650 Amber Lin: So, that tells us that

127 00:43:09.240 00:43:24.579 Amber Lin: okay, seems like VP customers like to get individual cookies instead of boxes, and but they don’t usually start off with… they usually start off with boxes. So that’s, like.

128 00:43:24.720 00:43:38.410 Amber Lin: That’s our interpretation of why things might… what it is, and then we can think about what the possibilities are of why that happened, of why do people usually get

129 00:43:38.710 00:43:40.800 Amber Lin: Boxes in the first order.

130 00:43:41.510 00:43:49.329 Amber Lin: And that… And with that question, if you look at their website, you’re gonna see that their free delivery

131 00:43:49.580 00:44:05.880 Amber Lin: limit is $20, so then you may make a good guess, is that, okay, maybe the first time I order on a website, I want to get a free delivery. So that’s why people… maybe that’s why people get boxes on their first order.

132 00:44:06.170 00:44:09.220 Amber Lin: But as they return, then…

133 00:44:09.250 00:44:26.319 Amber Lin: They might get more individual cookies because they know what they want, and they don’t want to spend that much, so they might just get the singular things that they like. And that’s the possible why of, this trend happening.

134 00:44:26.900 00:44:35.189 Amber Lin: So, I think that’s where we add these callouts of what we think things are happening, and we add… I add the title.

135 00:44:35.320 00:44:36.150 Amber Lin: There.

136 00:44:37.080 00:44:43.380 Amber Lin: And then the next part is making the recommendation for the clients of, okay, why…

137 00:44:43.700 00:45:02.420 Amber Lin: yes, this is happening, but what can we do about it? What does it mean for me? How can I… essentially, how can I make… how can a client make for money, or stop, losing money in some places? And this… in this scenario.

138 00:45:03.360 00:45:07.410 Amber Lin: Because boxes are higher value.

139 00:45:07.890 00:45:15.880 Amber Lin: the client would want people to order more boxes, right? So that’s one of the house of recommendations we can make.

140 00:45:16.200 00:45:17.489 Holly Condos: Oh, that’s item.

141 00:45:19.110 00:45:20.000 Amber Lin: Hello?

142 00:45:21.220 00:45:23.420 Amber Lin: Okay, I thought someone was talking.

143 00:45:23.650 00:45:27.059 Amber Lin: So we can either help them

144 00:45:27.610 00:45:32.650 Amber Lin: Think of solutions to sell more boxes to the returning customers.

145 00:45:32.710 00:45:47.990 Amber Lin: Or, maybe the trend of people getting more cookie… individual cookies on their return purchases, so when they come back, they like to get more individual cookies. Maybe that’s something we can use to get more people to return.

146 00:45:48.120 00:45:53.800 Amber Lin: So that could tell us what we can potentially market

147 00:45:54.230 00:45:56.950 Amber Lin: To those… to get people to return.

148 00:45:57.160 00:46:05.029 Amber Lin: So those are recommendations based on what we found, and then based on that, what can the clients do to…

149 00:46:05.150 00:46:07.170 Amber Lin: earn more money, essentially.

150 00:46:09.420 00:46:20.040 Amber Lin: Okay, and then the last part is to pick from all the insights something that has a through line of… in this case, I was talking about

151 00:46:20.090 00:46:31.160 Amber Lin: the life cycle, essentially, is when people make their first cookie purchase, and then when they return and make their second, third, fourth. So everything I put here is about

152 00:46:31.160 00:46:42.059 Amber Lin: one thing. And that just helps the client’s brain to digest it together, because this is a lot of information for people to digest, so we gotta pick the ones that

153 00:46:43.170 00:46:58.609 Amber Lin: have something in common, and then tells a progressive story, is that it builds off of each other, and there’s a, because of this, we did that, and because of that, we found out this. So there’s a logical progression between the slides.

154 00:46:58.610 00:47:06.659 Amber Lin: So that someone that’s listening to your story can actually understand that, like, something is happening progressively.

155 00:47:08.180 00:47:15.730 Amber Lin: So that’s the end of my presentation. Any questions about

156 00:47:16.110 00:47:22.689 Amber Lin: how you can do this, or any examples you have questions on, you can always message me as well.

157 00:47:25.910 00:47:28.320 Hannah Wang: Yeah, I know we’re on ti- er…

158 00:47:28.320 00:47:31.020 Amber Lin: We’re… the meeting’s over, but…

159 00:47:31.070 00:47:36.589 Hannah Wang: Yeah, I have time, but I have a question. I’ll message you later.

160 00:47:36.590 00:47:37.210 Amber Lin: Okay.

161 00:47:37.500 00:47:49.759 Amber Lin: Sounds good. I know more and more people are trying to do analysis, so hopefully this helps and this is timely. And always let me know if you have any questions, because I think

162 00:47:49.930 00:47:55.680 Amber Lin: I’ve spent most of my time

163 00:47:55.800 00:47:57.770 Amber Lin: making slides, and I made it.

164 00:47:58.320 00:47:59.730 Amber Lin: A lot of slides.

165 00:47:59.930 00:48:09.809 Amber Lin: So let me know if you have any questions about what to put on them, how to do the story, and then how to do the recommendations.

166 00:48:11.730 00:48:13.559 Amber Lin: Okay, thank you. Your slides are pretty.

167 00:48:13.560 00:48:14.640 Robert Tseng: Pretty good now, Amber.

168 00:48:14.940 00:48:25.709 Amber Lin: Yeah, thank you, Robert. I’ve been… I’ve been learning, so it’s… a lot of my knowledge has came from Robert, so if Robert does not have time, we can think about it together.

169 00:48:25.820 00:48:30.430 Amber Lin: So, another point of resource for everybody.

170 00:48:33.260 00:48:38.639 Amber Lin: All right. Thanks, everybody. Thanks for staying on the call. Thanks for taking time today.

171 00:48:38.640 00:48:39.150 Henry Zhao: Dude.

172 00:48:39.150 00:48:44.300 Amber Lin: We’ll see what Utam has to say after his ABC visit.

173 00:48:45.340 00:48:46.740 Robert Tseng: Yeah. Okay.

174 00:48:46.740 00:48:48.090 Amber Lin: Alright, bye everyone.

175 00:48:48.090 00:48:49.540 Hannah Wang: Oh, bye.