Meeting Title: Brainforge AI Team Sync Date: 2026-02-03 Meeting participants: Samuel Roberts, Uttam Kumaran
WEBVTT
1 00:00:06.560 ⇒ 00:00:07.620 Uttam Kumaran: Eddie.
2 00:00:13.990 ⇒ 00:00:14.950 Uttam Kumaran: Can you hear me?
3 00:00:19.210 ⇒ 00:00:20.470 Uttam Kumaran: Huh? Okay.
4 00:00:20.950 ⇒ 00:00:22.469 Uttam Kumaran: Okay, I gotcha. Yes.
5 00:00:23.510 ⇒ 00:00:25.150 Samuel Roberts: Weird, okay, that’s really weird.
6 00:00:25.150 ⇒ 00:00:27.850 Uttam Kumaran: Dude, I’m gonna buy you a new… a new mic, bro.
7 00:00:27.850 ⇒ 00:00:31.160 Samuel Roberts: I was gonna say, I’ve been thinking about it, because, like, something is whack with my.
8 00:00:31.160 ⇒ 00:00:34.240 Uttam Kumaran: Wait, tell me what you need! Tell me what you need, we’ll get it for you.
9 00:00:34.240 ⇒ 00:00:35.209 Samuel Roberts: Yeah, I’m thinking.
10 00:00:35.210 ⇒ 00:00:37.679 Uttam Kumaran: We actually have a new expense process.
11 00:00:37.680 ⇒ 00:00:41.609 Samuel Roberts: Yeah, I saw that, that gave me some thought about it, because, like, the camera works.
12 00:00:41.900 ⇒ 00:00:43.779 Samuel Roberts: But the mic and the camera?
13 00:00:45.260 ⇒ 00:00:53.190 Uttam Kumaran: That’s the thing, like, so I use… so I use this, this Jabra, I like because it’s actually incredibly sensitive.
14 00:00:53.190 ⇒ 00:00:54.460 Samuel Roberts: Yeah. And…
15 00:00:54.460 ⇒ 00:01:01.450 Uttam Kumaran: it’s like, you can sort of put it on, like, kind of take it off. It’s kind of… I mean, like, whatever. I use this if I’m, like.
16 00:01:01.590 ⇒ 00:01:03.829 Uttam Kumaran: I have su- I really just have some, like.
17 00:01:04.129 ⇒ 00:01:13.729 Uttam Kumaran: time today to focus, so then if I’m focusing, I put these on, and then I have my AirPods, basically. And then if I don’t have anything, I have the mic on this camera, so…
18 00:01:13.730 ⇒ 00:01:31.330 Samuel Roberts: So the mic on the camera not only doesn’t work all the time on this, but literally FaceTime, I have to open up the MIDI settings and change the, like, frequency back and forth every time my parents call me on that, because it’s worth every time my parents call me on that, because they’re the only people that call me on FaceTime. Yeah, a mic is probably the best bet here, because…
19 00:01:31.820 ⇒ 00:01:32.280 Uttam Kumaran: Yeah.
20 00:01:32.280 ⇒ 00:01:33.390 Samuel Roberts: Do some research.
21 00:01:33.390 ⇒ 00:01:40.230 Uttam Kumaran: Yeah, do some research, even if, like, this is probably the easiest to get. If you want, also, like, a flat mic, like.
22 00:01:40.460 ⇒ 00:01:49.839 Uttam Kumaran: like a Blue Yeti or something, we could do that. It’s up to you. I like my headphones, I don’t know if I wanna… Okay, okay. Then this is good, because this you could wear for a long time, and like, I’ll see…
23 00:01:49.840 ⇒ 00:01:51.649 Samuel Roberts: I like my headphones, is what I’m saying.
24 00:01:51.650 ⇒ 00:01:52.760 Uttam Kumaran: Oh, okay, okay.
25 00:01:52.760 ⇒ 00:01:55.970 Samuel Roberts: So I think, like… What headphones do you have?
26 00:01:56.770 ⇒ 00:01:57.980 Uttam Kumaran: Which headphones do you have?
27 00:01:57.980 ⇒ 00:02:02.309 Samuel Roberts: Oh my god, these are the Bose QuietComfort. They’re, like, over a decade old at this point.
28 00:02:02.310 ⇒ 00:02:07.689 Uttam Kumaran: Oh, yeah, dude, why… well then, why don’t you… I… we could just get you the Bose, like, QC… I forgot what it is.
29 00:02:07.690 ⇒ 00:02:14.309 Samuel Roberts: Maybe that’s too bad. I don’t know. These are still working, because I’ve looked at that a few times. The thing about it is, if I use the mic.
30 00:02:14.490 ⇒ 00:02:24.870 Samuel Roberts: On the headphones, it uses one channel, but if I use the mic on the camera, it uses two, so I get better audio quality. Oh. Well, the other thing is, these Sonys…
31 00:02:24.870 ⇒ 00:02:27.470 Uttam Kumaran: They… they switch pretty well.
32 00:02:27.700 ⇒ 00:02:28.569 Samuel Roberts: Oh, nice, okay.
33 00:02:28.570 ⇒ 00:02:30.850 Uttam Kumaran: You can connect to two devices at the same time.
34 00:02:30.850 ⇒ 00:02:33.569 Samuel Roberts: See, that’s the other problem, is my phone clicks onto the hood.
35 00:02:33.570 ⇒ 00:02:34.340 Uttam Kumaran: Yeah.
36 00:02:34.340 ⇒ 00:02:44.289 Samuel Roberts: I have to disconnect it every time, because if we’re talking here, and I have my headphones on, and a call comes in, it just hijacks the headphone. It’s a whole setup thing. I’ll do some… I’ll do some thinking about what I need.
37 00:02:44.290 ⇒ 00:02:46.739 Uttam Kumaran: Okay, do some, yeah, do some thinking, and yeah.
38 00:02:46.740 ⇒ 00:02:52.829 Samuel Roberts: Yeah, ugh. Anyway, yeah, I saw your message, I hadn’t gotten to dig and reply yet, but yes, I think…
39 00:02:52.990 ⇒ 00:02:54.110 Samuel Roberts: Couple things.
40 00:02:54.250 ⇒ 00:02:59.459 Samuel Roberts: Yeah, I realized, Lilo-wise, I got pulled in with the MCP stuff when Pranat was out, and I kinda haven’t…
41 00:02:59.670 ⇒ 00:03:00.190 Samuel Roberts: Extra kitty.
42 00:03:00.190 ⇒ 00:03:03.039 Uttam Kumaran: Got it out. Yeah. Which is dangerous.
43 00:03:03.040 ⇒ 00:03:04.289 Samuel Roberts: It’s okay, but…
44 00:03:04.290 ⇒ 00:03:04.969 Uttam Kumaran: Yeah, yeah, yeah.
45 00:03:04.970 ⇒ 00:03:09.360 Samuel Roberts: I need to back up a little bit. I think we will now that things are a little more stable on the platform there, so…
46 00:03:09.360 ⇒ 00:03:22.889 Uttam Kumaran: Yeah, and you have the… I think you have the luxury a little bit of, like, just having these two clients. Yeah. But I actually think that that’s, like, that’s an indication of, like, okay, we need to find ways to fill the book a bit more, so…
47 00:03:22.890 ⇒ 00:03:23.390 Samuel Roberts: Definitely.
48 00:03:23.390 ⇒ 00:03:27.719 Uttam Kumaran: put some… yeah, maybe we could go… we could go line by line. I just woke up today, I was like.
49 00:03:27.950 ⇒ 00:03:32.800 Uttam Kumaran: okay, let’s… let me just… so I’m sort of, like, going down a list of, like, different parts of the business. I’m like, okay.
50 00:03:32.800 ⇒ 00:03:33.690 Samuel Roberts: Yeah, totally.
51 00:03:33.690 ⇒ 00:03:42.959 Uttam Kumaran: I’ve been working on platform stuff now, I’m like, damn, I need, like, a shit ton of help. And so, yeah, maybe we could go line by line, you could tell me what your thoughts are.
52 00:03:42.960 ⇒ 00:03:44.960 Samuel Roberts: Yeah, yeah, okay, so, yeah, I think…
53 00:03:45.340 ⇒ 00:03:53.000 Samuel Roberts: I think I definitely, yeah, I need to step back a little bit on Lilo. There’ll probably be a little bit of things I can jump in here and there, but I’m glad… I was glad I got the thing set up.
54 00:03:53.180 ⇒ 00:04:06.459 Samuel Roberts: with, like, the MCPs and everything, because that’s going to scale better than if we had just done the environment variables and things, so… I’m glad I was there for that part, and then getting them running. But at this point, it’s pretty good, so I’m sure I can…
55 00:04:06.680 ⇒ 00:04:08.969 Samuel Roberts: Be a little more of that architect-planner role.
56 00:04:11.520 ⇒ 00:04:15.260 Samuel Roberts: Yeah, especially now, like, that Pranav’s been doing most of the work that’s, like.
57 00:04:16.060 ⇒ 00:04:20.110 Samuel Roberts: this POC. A lot of the stuff I’ve been doing has been, like, little cleanup things.
58 00:04:21.019 ⇒ 00:04:32.300 Samuel Roberts: From the other stuff, so I think it’s… I think we’re in a good place for that. I definitely need to do it, because yeah, I… I was thinking about this, too, because talking with Awasht during the service leads meeting, you know, he’s a little…
59 00:04:32.730 ⇒ 00:04:40.879 Samuel Roberts: pulled in more directions and not on every project he’s on. So, like, if my linear boards aren’t fully up to date, I know what’s going on, though, because I’m.
60 00:04:40.880 ⇒ 00:04:41.570 Uttam Kumaran: Yeah.
61 00:04:41.570 ⇒ 00:04:54.529 Samuel Roberts: And so, we’ve been talking about that a little bit, about, you know, I have that luxury right now, and I need to make sure that we’re doing it right now, so as we have more clients, it’s… I’m not as plugged in, I’ll still know what’s happening. Yes. So I definitely feel that, and that’s been something I’ve been thinking about a lot.
62 00:04:54.880 ⇒ 00:04:55.530 Uttam Kumaran: Okay, okay.
63 00:04:55.530 ⇒ 00:05:00.390 Samuel Roberts: But yeah, I think as… and that’s the other thing, like, just filling some time sometimes, is like, I can grab that Pranab, you know, it’s like, it’s…
64 00:05:00.650 ⇒ 00:05:05.720 Samuel Roberts: especially, like, with, you know, cursor half the time, I’m just like, I know exactly how it needs to get done, and I can just give it to.
65 00:05:05.720 ⇒ 00:05:12.919 Uttam Kumaran: Well, that’s also… that’s also my broader point, is that I want me, you, and Awash to turn more into, like, architects.
66 00:05:12.920 ⇒ 00:05:13.330 Samuel Roberts: Yeah.
67 00:05:13.330 ⇒ 00:05:14.400 Uttam Kumaran: and plan.
68 00:05:14.500 ⇒ 00:05:25.339 Uttam Kumaran: And start to… and then hand off once… once we decide the plan. Because what we’re good at is, like, knowing all the edge cases, and then we can hand off to people to just, like, yo, go make sure this plan
69 00:05:25.630 ⇒ 00:05:32.590 Uttam Kumaran: Gets done, and so we’re actually accomplishing what we want in a lot… less time.
70 00:05:32.590 ⇒ 00:05:34.990 Samuel Roberts: Yeah, the only thing there is, like.
71 00:05:35.270 ⇒ 00:05:44.689 Samuel Roberts: sometimes, if I have a… like, I can chat with Cursor about a plan, and it can do it, like, a decent chunk of the time. The thing is, like, I probably need to be… get to a good point where I can, like.
72 00:05:46.180 ⇒ 00:05:50.500 Uttam Kumaran: Or you just need to find where… yeah, you need to… or you need to time box it or something.
73 00:05:50.500 ⇒ 00:05:51.960 Samuel Roberts: That’s my other issue, is I’ll get pulled.
74 00:05:51.960 ⇒ 00:05:59.919 Uttam Kumaran: It’s easy to get… it’s easy to get there, and then you’re… you’re doing… then you’re doing work that, like, dude, you’re way… you’re, like, overqualified for, which is just, like.
75 00:05:59.920 ⇒ 00:06:13.300 Uttam Kumaran: the iteration. So, yes, if you’re, like, write the plan, have it do the first one, ship the draft PR, I’m… I don’t… yeah, I don’t mind, but, like, I think what I… what I think you’re gonna have the challenge of, just because I know you, is the, like.
76 00:06:13.320 ⇒ 00:06:16.209 Uttam Kumaran: Getting out of the sinkhole.
77 00:06:16.210 ⇒ 00:06:16.540 Samuel Roberts: Yeah.
78 00:06:16.540 ⇒ 00:06:31.270 Uttam Kumaran: And so, tell me… tell me or tell us how to help you do that, because I certainly… one good way I find is, like, I’m starting to just block… time block my schedule to do certain things, but in particular, I really… I think…
79 00:06:31.340 ⇒ 00:06:43.619 Uttam Kumaran: where I want to see some more time in the week help from you is, one, certainly on, like, almost coming alongside me on the platform, and trying to be like, how do I enable
80 00:06:43.920 ⇒ 00:06:56.390 Uttam Kumaran: this person who’s trying to ship all these things to do that in a much more safe manner. And then also, like, you’re… you also know how many things we want to build, so you could also, like… dude, I’m building stuff, like, in 30 minutes.
81 00:06:56.390 ⇒ 00:06:56.950 Samuel Roberts: Yeah.
82 00:06:56.950 ⇒ 00:07:00.679 Uttam Kumaran: You can also fi- you’re gonna be… it’ll be way faster for you, so…
83 00:07:00.830 ⇒ 00:07:16.589 Uttam Kumaran: part of that is, like, how do me and you spend more time on that each week, so that we can, like, really be like, okay, we need to set up our codex environment in a certain way, we need to modify the… like, I want to have more conversations about the ergonomics of vibe coding around here.
84 00:07:17.080 ⇒ 00:07:32.739 Uttam Kumaran: You know, and then the second piece is, like… the second and third pieces are, one, like, the go-to-market team is gonna need some of your time as well, on case studies and on helping them articulate, like, what it is we’re doing for clients. You’re the only one on the AI team
85 00:07:32.930 ⇒ 00:07:34.629 Uttam Kumaran: With the time and, like, the…
86 00:07:34.970 ⇒ 00:07:37.310 Uttam Kumaran: Elevation to, like, be able to do that.
87 00:07:37.420 ⇒ 00:07:43.210 Uttam Kumaran: I’m filling some of that on, but the problem… on the data side, we have a long history of doing data work here, so…
88 00:07:43.280 ⇒ 00:07:43.910 Samuel Roberts: Right.
89 00:07:43.910 ⇒ 00:07:52.430 Uttam Kumaran: they’re not gonna have a challenge there. On the AI side, I think… and the best way to do that is actually just, like, if you just carve out 2 hours to spend with Luke.
90 00:07:52.770 ⇒ 00:07:55.789 Uttam Kumaran: That’s, like, basically all you need to do to close that one.
91 00:07:56.020 ⇒ 00:07:56.630 Samuel Roberts: Okay.
92 00:07:56.630 ⇒ 00:07:58.989 Uttam Kumaran: Because he’ll come in with a ton of questions for you.
93 00:07:58.990 ⇒ 00:08:00.040 Samuel Roberts: And…
94 00:08:00.460 ⇒ 00:08:09.719 Uttam Kumaran: That would be great. And then the last piece is on recruitment. And this is where I haven’t been more explicit, but, like, we’re… we’ll be presenting on Friday of, like, our recruitment plan.
95 00:08:09.720 ⇒ 00:08:10.130 Samuel Roberts: Okay.
96 00:08:10.130 ⇒ 00:08:17.230 Uttam Kumaran: But basically, I’m gonna be going with… I was gonna give this feedback to Awash as well, but…
97 00:08:17.440 ⇒ 00:08:27.879 Uttam Kumaran: I’ve been doing a lot of the first rounds. I’m gonna still do some first rounds for, like, people that are, like, very, very senior. Like, for example, if I… when I called you, if I just was like.
98 00:08:28.110 ⇒ 00:08:32.460 Uttam Kumaran: go into this process. It may not have, like.
99 00:08:32.590 ⇒ 00:08:36.090 Uttam Kumaran: You wouldn’t have seen me and been like, okay, there’s, like, this is, like, a legitimate thing.
100 00:08:36.090 ⇒ 00:08:36.989 Samuel Roberts: Right, right, right.
101 00:08:36.990 ⇒ 00:08:46.959 Uttam Kumaran: For some people, I still want to spend time doing that, because it matters that we try to give a great shot. For some people, I just want to throw them into flow, so there’s gonna be a first round, like, screening.
102 00:08:47.090 ⇒ 00:08:49.820 Samuel Roberts: Okay. And so that’s where I need some help, and then…
103 00:08:49.820 ⇒ 00:08:52.709 Uttam Kumaran: You’re the fir- you’re gonna be the first interview for… for them.
104 00:08:52.710 ⇒ 00:08:54.499 Samuel Roberts: Okay, yeah, that sounds great.
105 00:08:54.500 ⇒ 00:09:00.330 Uttam Kumaran: It’s gonna be you, then Pranav, and then we’re gonna go into the panel for presentation on, like, the…
106 00:09:00.610 ⇒ 00:09:04.100 Uttam Kumaran: whatever solution that we have, I think. Cool. Yeah. AI solution.
107 00:09:04.320 ⇒ 00:09:04.960 Samuel Roberts: Sweet.
108 00:09:05.380 ⇒ 00:09:09.449 Samuel Roberts: Yeah, no, that sounds good. I’ve been… I was a little… I was wondering a little bit about some of that. I wasn’t sure…
109 00:09:09.620 ⇒ 00:09:10.110 Samuel Roberts: where…
110 00:09:10.110 ⇒ 00:09:19.190 Uttam Kumaran: That one, I haven’t… I’ve been just taking on, and again, but this is where, like, I… I… I want you to look at that list of priorities and be like, how…
111 00:09:19.340 ⇒ 00:09:28.420 Uttam Kumaran: can I enable you to accomplish those? Like, do we want to, like, set up fixed meetings to just be like, yo, for an hour, we’re gonna talk about recruitment each week?
112 00:09:28.420 ⇒ 00:09:28.930 Samuel Roberts: Yeah, maybe.
113 00:09:28.930 ⇒ 00:09:47.869 Uttam Kumaran: That’s great. If you feel comfortable blocking out your time to doing that, that’s fine. But this is where, like, consider that, like, me, Clarence, the SL team, our… the ops team are all tools for you to use to accomplish these, right? You know, versus… I don’t… this… you’re not on your own being like, okay, I gotta, like, balance all these four things.
114 00:09:47.870 ⇒ 00:10:02.370 Samuel Roberts: Yeah, that’s definitely, that’s definitely something I felt a little bit, trying to… and it was even interesting talking with, with, with Clarence during the SL meeting yesterday. We were talking about CloudBot, or OpenClaw now, and he was like, it’s very similar to, like, the leadership-type skills you need to work on, because you’re, like.
115 00:10:02.590 ⇒ 00:10:13.280 Samuel Roberts: delegating to these agents, and, like, the specificity, and, like, he was talking, and I was like, he made a really… he, like, brought it full circle to what we were talking about. I was like, yeah, I gotta think more about the resources I have, and…
116 00:10:13.540 ⇒ 00:10:15.460 Samuel Roberts: Not just delegating, but, like.
117 00:10:15.460 ⇒ 00:10:18.260 Uttam Kumaran: Yeah. Pulling the numbers the right way, because I think sometimes I get…
118 00:10:18.260 ⇒ 00:10:20.129 Samuel Roberts: Nerd sniped into it.
119 00:10:20.130 ⇒ 00:10:27.149 Uttam Kumaran: Yes, yes. No, that’s exactly it, like, I… so, it’s been the story of my job here to not do that.
120 00:10:27.150 ⇒ 00:10:27.600 Samuel Roberts: Yeah.
121 00:10:27.600 ⇒ 00:10:33.989 Uttam Kumaran: But that’s the thing, is I just don’t do much stuff anymore that’s, like, really fun, so I’m always trying to finish it and move on to the next thing.
122 00:10:33.990 ⇒ 00:10:34.580 Samuel Roberts: Totally.
123 00:10:34.580 ⇒ 00:10:44.420 Uttam Kumaran: But I rea… like, I… I think I tend to agree in that, like, what… what… I think what you should be thinking about is the fact that you’re so involved with Lilo should be…
124 00:10:44.420 ⇒ 00:10:44.960 Samuel Roberts: so involved.
125 00:10:44.960 ⇒ 00:10:53.070 Uttam Kumaran: a side effect of us not enabling them to, like… not enabling an AI-driven Development process, right?
126 00:10:53.070 ⇒ 00:10:53.590 Samuel Roberts: Yeah.
127 00:10:53.590 ⇒ 00:11:05.480 Uttam Kumaran: Because if you were to… like, I’m almost like, okay, if we were to go back and be like, half the features you pushed end-to-end, if you were to develop the plan in 10 minutes.
128 00:11:06.030 ⇒ 00:11:10.200 Uttam Kumaran: kicked it off and been like, yo, when this is done, you take it from there.
129 00:11:10.510 ⇒ 00:11:14.109 Uttam Kumaran: what… how would that have changed? And so, part of this is, like.
130 00:11:15.600 ⇒ 00:11:20.860 Uttam Kumaran: The fact that your time is getting sucked there should be an indication that we’re not doing enough.
131 00:11:21.230 ⇒ 00:11:24.259 Uttam Kumaran: You know, on the AI spread development side.
132 00:11:24.280 ⇒ 00:11:26.149 Samuel Roberts: That’s… that’s… that’s good.
133 00:11:26.150 ⇒ 00:11:26.840 Uttam Kumaran: Yeah.
134 00:11:27.960 ⇒ 00:11:28.580 Samuel Roberts: Okay.
135 00:11:30.390 ⇒ 00:11:31.540 Uttam Kumaran: I don’t know, what do you think?
136 00:11:31.770 ⇒ 00:11:47.049 Samuel Roberts: Yeah, no, I think that’s true. I think part of it was, you know, it was a few things, you know, it was definitely, like, I had to do some of that dev work early and just getting myself back out of it once it’s like, oh, I have that, let me just fix that kind of thing. It’s definitely, like, working with Pranav to start, I wasn’t sure.
137 00:11:47.650 ⇒ 00:11:50.820 Samuel Roberts: to throw. I feel a little more comfortable with that now.
138 00:11:51.770 ⇒ 00:11:55.889 Samuel Roberts: In case, oh yeah, I’m not worried about throwing stuff in the case. I kind of know what people can handle now, so…
139 00:11:56.580 ⇒ 00:11:58.710 Samuel Roberts: But yeah, I think… I think it definitely…
140 00:11:59.130 ⇒ 00:12:01.479 Samuel Roberts: Taking that step back, taking, like, the first swing.
141 00:12:02.160 ⇒ 00:12:07.490 Samuel Roberts: Like, get a plan out, be like, yeah, run with it, you know, ask me whatever questions, like, clarify.
142 00:12:07.490 ⇒ 00:12:10.980 Uttam Kumaran: But that’s also the thing, is that me and you are the only people at that level.
143 00:12:10.980 ⇒ 00:12:11.580 Samuel Roberts: Yeah.
144 00:12:11.580 ⇒ 00:12:14.620 Uttam Kumaran: So, like, if I don’t… if we don’t talk to each other.
145 00:12:14.810 ⇒ 00:12:15.300 Samuel Roberts: True.
146 00:12:15.300 ⇒ 00:12:19.890 Uttam Kumaran: and figure out, like, how this works, you’re gonna be on your own, and I’m gonna be on my own.
147 00:12:19.890 ⇒ 00:12:20.610 Samuel Roberts: No, you’re right.
148 00:12:20.610 ⇒ 00:12:23.359 Uttam Kumaran: No one else in the company that is thinking about
149 00:12:23.710 ⇒ 00:12:28.830 Uttam Kumaran: It’s like, it’s like, actually both reading about all the stuff that’s happening.
150 00:12:28.850 ⇒ 00:12:47.669 Uttam Kumaran: And being like, how do we adopt it, right? So, that’s where, like, we just… I want to just spend more time, because, like, Gabe is okay, but, like, I’m trying to… he’s just clearly, like, not technical enough, and he’s coming at it from just, like, he’s really smart, but I instead need him… I want him to go focus more on adoption.
151 00:12:47.820 ⇒ 00:12:56.310 Uttam Kumaran: Which is, like, take everything we’ve built, and continue to build, and just make sure you go teach people. And we’ll teach you, and then you go teach everybody, kind of deal.
152 00:12:56.560 ⇒ 00:13:03.180 Uttam Kumaran: So, for me, it’s the same thing. It’s like, you sent me something about Conductor. I also… I’ve been following Conductor for, like, a month.
153 00:13:03.510 ⇒ 00:13:04.170 Samuel Roberts: Yeah.
154 00:13:04.170 ⇒ 00:13:21.539 Uttam Kumaran: Yeah, finally I got to the point where I was like, I’m ready, I’m working on too many things to be able to use WorkTrees. Work trees can only use single repo. So then I was like, oh shit, I forgot about Conductor. I, like, looked through my bookmarks for Conductor, and then I started using it. But, like, again, I… I’m just gonna be doing it
155 00:13:21.710 ⇒ 00:13:28.730 Uttam Kumaran: like, alone, randomly. So, like, we should just talk and figure it out, because then, ultimately, I want to roll out Conductor to…
156 00:13:28.970 ⇒ 00:13:30.610 Uttam Kumaran: Everybody that can use it, you know?
157 00:13:30.610 ⇒ 00:13:32.299 Samuel Roberts: Did you look at the Codex app, too?
158 00:13:32.590 ⇒ 00:13:34.579 Uttam Kumaran: I… I have to install it today, yeah.
159 00:13:34.580 ⇒ 00:13:45.230 Samuel Roberts: Yeah, because you had mentioned Conductor, and then I was watching a few videos about it, and someone mentioned that it’s, like, it’s basically a GUI on top of the Codex CLI, kind of the way Conductor is for… and I was like, oh, but then…
160 00:13:45.600 ⇒ 00:13:48.729 Samuel Roberts: I still like cursor, because I like being able to change the models and…
161 00:13:48.950 ⇒ 00:14:00.099 Samuel Roberts: you know, I don’t want to be sucked into one ecosystem necessarily, but, I think they… the UI looks really nice, and, like, the flow, I think it’s what Cursor’s getting towards a little bit, but not quite there yet.
162 00:14:01.950 ⇒ 00:14:09.880 Samuel Roberts: Yeah. So I was gonna try to play with that a little bit today. And the nice thing about Codex is it works… you can do local, work tree, and cloud.
163 00:14:11.380 ⇒ 00:14:12.989 Uttam Kumaran: Oh, really? Okay.
164 00:14:12.990 ⇒ 00:14:15.080 Samuel Roberts: kick off jobs, and I’m wondering if we can, like…
165 00:14:15.380 ⇒ 00:14:19.220 Samuel Roberts: Kick off a job and pass it along to someone else to, like, see through on the.
166 00:14:19.220 ⇒ 00:14:26.269 Uttam Kumaran: Well, another hack that I was reading on Twitter is some people are using Opus to kick off codec CLIs.
167 00:14:26.640 ⇒ 00:14:29.030 Samuel Roberts: Of course they are, oh my gosh, yes.
168 00:14:29.030 ⇒ 00:14:34.210 Uttam Kumaran: You know? So, like, it’s more conducting, and so this is where, like, we just have to play around, because the other…
169 00:14:34.210 ⇒ 00:14:34.719 Samuel Roberts: Good thing, dude.
170 00:14:34.720 ⇒ 00:14:41.530 Uttam Kumaran: is, like, I… the data team is not using this shit enough. Okay. Because there is a bit of a gap of, like, doing data work and, like.
171 00:14:41.530 ⇒ 00:14:42.109 Samuel Roberts: I use this stuff.
172 00:14:42.110 ⇒ 00:14:45.309 Uttam Kumaran: So, so, again, like, I just need, like.
173 00:14:45.840 ⇒ 00:14:49.770 Uttam Kumaran: 10 hours of your time to go to be like, how the fuck does this company, like.
174 00:14:49.770 ⇒ 00:14:52.169 Samuel Roberts: ratchet up its AI use. Totally.
175 00:14:52.170 ⇒ 00:15:11.309 Uttam Kumaran: And… and Robert is all in on the sales side. Like, can I get… can I show you something I just did today, which is actually so legit, because Robert set this up? So, Plan Medicare got back to us being like, oh, this is, like, too expensive, and so Robert actually set up a bunch of things in Vault on objection handling.
176 00:15:11.450 ⇒ 00:15:17.580 Uttam Kumaran: And so, I sent it the email. It said… it asked me some follow-up questions.
177 00:15:17.580 ⇒ 00:15:17.970 Samuel Roberts: Like, for example.
178 00:15:17.970 ⇒ 00:15:19.769 Uttam Kumaran: For example, Brian is pushing back.
179 00:15:20.090 ⇒ 00:15:34.409 Uttam Kumaran: this is a sales objection handling. I should let the user think through the subscriptions. And it went through, and it looked through, like, price objection, competitive threat, value alignment, opportunity cost. Robert wrote all of this in the go-to-market thing about how to do
180 00:15:34.610 ⇒ 00:15:36.850 Uttam Kumaran: Price, how to do objection handling.
181 00:15:36.850 ⇒ 00:15:37.590 Samuel Roberts: Yeah.
182 00:15:37.590 ⇒ 00:15:42.279 Uttam Kumaran: And it then asked me some questions on how I want to handle it, and then it drafted me, and I ripped it.
183 00:15:42.520 ⇒ 00:15:47.830 Samuel Roberts: Nice. Yeah, I saw that reply in the email, and I was like, I… I… it was so fast. I had to be like…
184 00:15:48.040 ⇒ 00:16:00.020 Uttam Kumaran: Yeah, but, like, dude, so this is this thing, is, like, every team, like, so only me and you, like, I have to show Robert the fact that this vault is possible for him to go, and so every team, we gotta think about this, so…
185 00:16:00.230 ⇒ 00:16:09.560 Uttam Kumaran: I think my ask for you today is, like, think about how I can help you do it. I’m telling you, whether it’s like, yo, in order for me to not get nerd site by stuff, can we just block out
186 00:16:09.610 ⇒ 00:16:24.410 Uttam Kumaran: an hour in the middle of the day, so I can… we can talk about this, and maybe it breaks it up. If you want to talk about recruitment, like, I’m happy to layer on the meetings until you get into the habit, and then take them out. So, you tell me how to be helpful.
187 00:16:24.600 ⇒ 00:16:33.529 Samuel Roberts: Alright, yeah, I’ll think about that a little bit today. I have some other stuff that I was digging into for, like, the monorepo and PR stuff, so, like, that’s… this is hopefully a good day when I can be a little more…
188 00:16:33.880 ⇒ 00:16:36.219 Samuel Roberts: Out of Lilo in this space.
189 00:16:36.500 ⇒ 00:16:47.159 Uttam Kumaran: Yeah, and I know me and you both are gonna be, like, we can develop really, really fast, and so part of this is also there’s gonna be some things where, like, we’re just gonna have to do, we can’t wait for other people to figure this out.
190 00:16:47.160 ⇒ 00:16:47.820 Samuel Roberts: Yeah.
191 00:16:47.820 ⇒ 00:16:54.460 Uttam Kumaran: And that’s what I’m showing you, right? Like, I’m able to push out features, like, I just pushed out a bunch of stuff for ops on the Brainforge Assistant last night.
192 00:16:54.580 ⇒ 00:16:55.699 Samuel Roberts: Yeah, I saw some of the links.
193 00:16:55.700 ⇒ 00:17:08.810 Uttam Kumaran: I’m able to… I don’t touch railway at all. I don’t touch, like, half the stuff anymore. It’s all CLI-driven. Nice. And so I wanted to take that and figure out how we scale that, so… I mean, just… just food for thought, so yeah, let’s…
194 00:17:08.819 ⇒ 00:17:09.419 Samuel Roberts: Absolutely.
195 00:17:09.420 ⇒ 00:17:12.859 Uttam Kumaran: If you have time today to catch up later, let’s… and you think about it, let me know.
196 00:17:12.869 ⇒ 00:17:18.429 Samuel Roberts: Yeah, I’ll definitely give some thought, and if we can catch up later, that’d be good. Okay. Okay, I gotta jump to the other one.
197 00:17:18.430 ⇒ 00:17:19.800 Uttam Kumaran: Okay, okay, okay, alright.
198 00:17:19.800 ⇒ 00:17:20.250 Samuel Roberts: Yep.