Meeting Title: AI in Legal Practice Discussion Date: 2026-02-24 Meeting participants: Robert Tseng, Lisa Temple
WEBVTT
1 00:01:02.810 ⇒ 00:01:05.459 Lisa Temple: Hi! How are you?
2 00:01:05.610 ⇒ 00:01:07.020 Robert Tseng: Hi, Lisa!
3 00:01:07.020 ⇒ 00:01:09.550 Lisa Temple: Hi, sorry I’m a minute late here.
4 00:01:09.550 ⇒ 00:01:10.560 Robert Tseng: No worries!
5 00:01:11.190 ⇒ 00:01:12.800 Robert Tseng: I’m good, how are you?
6 00:01:13.440 ⇒ 00:01:18.719 Lisa Temple: Good, good. I’ve heard so many nice things about you, and I’m so happy to connect.
7 00:01:19.010 ⇒ 00:01:25.969 Robert Tseng: Oh, alright, I don’t know what John shared with you, but, yeah, no, I appreciate him a lot, and…
8 00:01:25.980 ⇒ 00:01:39.150 Robert Tseng: Yeah, it’s… we’ve been in touch for the past couple years, so it’s crazy how, like, before, I wasn’t… I mean, I’m… I don’t know if he told you, but I… I’m in, I’m currently at Fordham, actually, right now, but, I’m a… I’m an evening law student here.
9 00:01:39.280 ⇒ 00:01:49.910 Robert Tseng: started that while I was still kind of, like, working on this company, so it’s been a… it’s been a really interesting past year, but, I love, like, kind of the trajectory of, like.
10 00:01:49.910 ⇒ 00:02:07.110 Robert Tseng: finding the intersection of, like, law, I guess data is my background in tech, and yeah, when he shared about… I didn’t know that you were a lawyer, and then I also didn’t know that you were, you were bringing so much AI into your practice, and so I was… I asked him if I can have a chat with you.
11 00:02:07.110 ⇒ 00:02:13.430 Lisa Temple: Oh, it’s my pleasure. Yeah, I’m pretty passionate about new technology, and of course, AI is
12 00:02:13.960 ⇒ 00:02:18.620 Lisa Temple: Revolutionized every field, so, including law.
13 00:02:19.010 ⇒ 00:02:19.420 Robert Tseng: Yeah.
14 00:02:19.420 ⇒ 00:02:23.229 Lisa Temple: It might be interesting, although I’m… we have a law…
15 00:02:23.360 ⇒ 00:02:33.740 Lisa Temple: a law clerk who’s a 3L at Quinnipiac, who’s working with us, and they don’t really use AI, in law school, so I was wondering if that’s…
16 00:02:34.190 ⇒ 00:02:35.940 Lisa Temple: The same for Fordham.
17 00:02:36.830 ⇒ 00:02:46.810 Robert Tseng: Yeah, I mean, I would say Fordham tries to be pretty, AI-forward, like, one of the early partners for Harvey AI, so I’ve been… I’ve been… I’ve had access to it and, like, looked around.
18 00:02:47.050 ⇒ 00:03:05.250 Robert Tseng: But yeah, obviously within classes, like, I think they try to… everything is switched over to closed book… book exams, and, they’re like, I don’t want to give people access to their computers, and we’re doing, like, the testing stuff, because I think people… they just assume that, you know, people are just going to cheat, so… which I understand, but,
19 00:03:05.250 ⇒ 00:03:05.690 Lisa Temple: Yeah.
20 00:03:05.690 ⇒ 00:03:16.419 Robert Tseng: Yeah, so it’s interesting, being in school at this time, and seeing how the rules are changing, like, semester to semester, and how the law school’s reacting to all of that, so…
21 00:03:16.420 ⇒ 00:03:18.780 Lisa Temple: Yeah, you, are you a 3L right now?
22 00:03:18.780 ⇒ 00:03:25.530 Robert Tseng: I’m not, I’m a… I’m a, I guess they call it a 1LE, like, I literally just started, like, this is my second semester, so…
23 00:03:25.530 ⇒ 00:03:27.739 Lisa Temple: Oh my gosh, that’s the hardest time.
24 00:03:28.540 ⇒ 00:03:30.589 Robert Tseng: Oh, yeah, yeah,
25 00:03:30.810 ⇒ 00:03:35.909 Robert Tseng: Yeah, I guess people told me it would be… it would be hard, but, you know, we’re getting through it.
26 00:03:36.170 ⇒ 00:03:41.260 Lisa Temple: And also, you’re doing it part-time, right? With… while you’re working?
27 00:03:41.370 ⇒ 00:03:43.050 Robert Tseng: Yeah. That’s, like, the hardest…
28 00:03:43.280 ⇒ 00:03:51.010 Lisa Temple: combination of life, I think. Besides having kids and, like, working. But yeah, that’s hard.
29 00:03:51.010 ⇒ 00:03:57.420 Robert Tseng: Yeah, I have a very understanding wife, and yeah, she’s been a good supporting system for me.
30 00:03:57.740 ⇒ 00:03:58.170 Lisa Temple: Terrific.
31 00:03:58.170 ⇒ 00:03:59.880 Robert Tseng: Yeah.
32 00:03:59.880 ⇒ 00:04:02.500 Lisa Temple: You’re lucky. That’s good. She’s a keeper.
33 00:04:02.760 ⇒ 00:04:03.090 Robert Tseng: Yeah.
34 00:04:04.900 ⇒ 00:04:15.679 Robert Tseng: Yeah, I guess you, you, so you’ve been running your own practice, I think John mentioned a few things, you have about, like, something around 10, 10 attorneys under you.
35 00:04:15.940 ⇒ 00:04:21.230 Lisa Temple: Oh, not 10 attorneys, 10 years, but… I have,
36 00:04:21.630 ⇒ 00:04:29.490 Lisa Temple: two full-time attorneys, and a number of other… I have the 3L, 3L.
37 00:04:29.490 ⇒ 00:04:30.100 Robert Tseng: Yeah.
38 00:04:30.100 ⇒ 00:04:35.060 Lisa Temple: We have contractors, and, a paralegal, and…
39 00:04:35.260 ⇒ 00:04:41.089 Lisa Temple: an executive assistant. So, it’s a… it’s a nice little group. It’s small, but we’re growing.
40 00:04:41.370 ⇒ 00:04:50.400 Lisa Temple: Yeah. So it’s been… a really exciting year, actually, with growth and with AI. So it’s,
41 00:04:50.820 ⇒ 00:05:07.559 Lisa Temple: it was kind of scary what the beginning of 2024… 5? Sorry, oh my gosh, I’m 2 years behind. The beginning of 2025 and into that year was slowing down, and then all of a sudden, it just shot right up. So it’s been a rollercoaster with both,
42 00:05:07.840 ⇒ 00:05:12.000 Lisa Temple: with learning how to work differently, mostly with AI and,
43 00:05:12.370 ⇒ 00:05:16.040 Lisa Temple: how to grow a business. I’m learning as well, I was never…
44 00:05:16.260 ⇒ 00:05:18.639 Lisa Temple: Like, I don’t really consider myself, like.
45 00:05:19.040 ⇒ 00:05:23.439 Lisa Temple: like John, who’s the leader of a business, I’ve just been…
46 00:05:23.440 ⇒ 00:05:23.920 Robert Tseng: Yeah.
47 00:05:23.920 ⇒ 00:05:26.150 Lisa Temple: a toiling attorney.
48 00:05:26.150 ⇒ 00:05:26.720 Robert Tseng: Hmm…
49 00:05:27.100 ⇒ 00:05:37.869 Lisa Temple: So, in fact, I am a leader of a business, it turns out, and I have to make decisions. So, you know, it’s been an exciting, year.
50 00:05:38.580 ⇒ 00:05:57.400 Robert Tseng: Yeah, I’d love to hear more about, kind of, how you’ve… AI has impacted your operating model. I guess, like, some more context, like, I’ve tried… I mean, we work with a wide range of businesses in my day job, like, I run, like, an AI data engineering consultancy, which I think has been really… I mean, I can share more about it later, but…
51 00:05:57.400 ⇒ 00:06:19.780 Robert Tseng: We’ve… we’ve tried talking to some, like, law firms as well. We’ve built a couple pilots, actually, just, like, interesting automation tools for them, but it doesn’t really, like, kind of, you know, there are things like, automated lead intake, and, like, I guess we worked with, like, a PI… personal injury firm before, and, like, helped them do, like, some automated, record retrieval. So, like, it’s… it’s more, like.
52 00:06:19.780 ⇒ 00:06:29.510 Robert Tseng: I mean, it was not… like, these are… to me, these are small projects that never really, like, sit… like, kind of transformed, like, their… their organization, and…
53 00:06:29.570 ⇒ 00:06:44.319 Robert Tseng: Yeah, I mean, I have opinions about, like, kind of what… like, why they didn’t go farther, but I’m… as somebody who’s really embraced it, and I’m so curious, like, kind of, how you thought about it, and what… how it’s impacting, kind of, your operating model now.
54 00:06:44.800 ⇒ 00:06:52.260 Lisa Temple: Oh, interesting. I want to hear more about your business, too. Yeah. So, to answer your question, though,
55 00:06:52.780 ⇒ 00:07:01.399 Lisa Temple: Maybe 3 years ago, we started looking into doing flat fees. We knew that,
56 00:07:01.400 ⇒ 00:07:01.720 Robert Tseng: Yeah.
57 00:07:01.720 ⇒ 00:07:16.599 Lisa Temple: AI was coming, and we kind of saw that it was here, really, but we weren’t using it that much. No law firm really was that I knew, because everyone was kind of scared of it.
58 00:07:17.270 ⇒ 00:07:20.710 Lisa Temple: but…
59 00:07:22.150 ⇒ 00:07:35.139 Lisa Temple: So… but we did realize that, you know, at some point, it was going to be front and center of what we do, and so, or how we work, rather. And so,
60 00:07:35.410 ⇒ 00:07:38.089 Lisa Temple: For the past few years, we’ve been…
61 00:07:38.300 ⇒ 00:07:51.300 Lisa Temple: really trying to find a better way to essentially bill our clients that works better for us, and for, sorry, my computer’s, like, doing this crazy update right now that I… I have no idea what’s happening.
62 00:07:51.900 ⇒ 00:07:59.989 Lisa Temple: So, I’m so sorry, things just keep popping up and distracting me, and I wanted to stop. Anyway,
63 00:08:00.760 ⇒ 00:08:07.269 Lisa Temple: So those two things have really been hand-in-hand for us, the, the pricing and the,
64 00:08:07.540 ⇒ 00:08:16.950 Lisa Temple: and AI, But more recently, and it felt a little like… what,
65 00:08:17.490 ⇒ 00:08:26.069 Lisa Temple: like, everyone was using ChatGPT, and they still are, but it felt, a little dangerous, even though if you subscribe to their…
66 00:08:26.720 ⇒ 00:08:30.630 Lisa Temple: we don’t share or use your information to train, yeah.
67 00:08:31.290 ⇒ 00:08:33.570 Lisa Temple: That felt better, but still, like…
68 00:08:34.299 ⇒ 00:08:41.799 Lisa Temple: Well, anyone can use it, so, like, why… you know, it… we use it as, like, kind of a check and make sure, like.
69 00:08:42.110 ⇒ 00:08:49.540 Lisa Temple: look… things looked buttoned up and help with emails and things, but didn’t feel totally comfortable using ChatGPT.
70 00:08:50.020 ⇒ 00:08:56.760 Lisa Temple: generally as a tool for legal work. But then, last year.
71 00:08:57.260 ⇒ 00:09:00.379 Lisa Temple: Some exciting new tools came on the market.
72 00:09:00.590 ⇒ 00:09:04.399 Lisa Temple: The two that we use most are,
73 00:09:05.060 ⇒ 00:09:12.150 Lisa Temple: maybe exclusively? I… we’d still use ChatGPT, but we use Clio Work. I don’t know if you’ve heard of it.
74 00:09:12.330 ⇒ 00:09:13.649 Robert Tseng: Yeah, I know Kia.
75 00:09:13.920 ⇒ 00:09:18.229 Lisa Temple: And then Law Insider, which we might use more than Clio Work.
76 00:09:19.080 ⇒ 00:09:22.460 Lisa Temple: We’re a Clio… we’re a Clio law firm.
77 00:09:22.600 ⇒ 00:09:23.760 Robert Tseng: Okay.
78 00:09:24.000 ⇒ 00:09:31.739 Lisa Temple: So, Clio work made sense to us, but it has made our Lexus AI subscription
79 00:09:32.130 ⇒ 00:09:36.520 Lisa Temple: moot, I want to say? And it’s so expensive, and it galls me.
80 00:09:36.670 ⇒ 00:09:43.970 Lisa Temple: So, anyway, that’s kind of the… I don’t know if I answered your question. Did I answer?
81 00:09:43.970 ⇒ 00:09:52.940 Robert Tseng: Yeah, yeah, no, thank you for sharing about the tools, and then also, like, kind of, like, how it’s forced you to kind of price differently. If I can kind of follow up on the pricing point, like, I’m curious, like.
82 00:09:52.940 ⇒ 00:10:17.570 Robert Tseng: how did you select which, like, you know, you have your… you have your… I mean, we’re a services company, too, so, like, I… I think about, like, hey, are there certain services that, like, as we continue to run them better, there’s more automation that we can put in place? Like, we could also just kind of start to make… turn into fixed fees and not really bill by the hour. So, I’m curious, like, how you, like, kind of made that… those… those decisions for… for yourself, like, what… what… what became fixed and what stays, like.
83 00:10:17.870 ⇒ 00:10:20.369 Robert Tseng: I guess hourly or variable.
84 00:10:20.980 ⇒ 00:10:21.850 Lisa Temple: Well…
85 00:10:22.930 ⇒ 00:10:31.839 Lisa Temple: We started thinking about flat fees around when a company called Altfee started coming into market, and we met them at ClioCon.
86 00:10:32.500 ⇒ 00:10:33.750 Lisa Temple: And…
87 00:10:33.850 ⇒ 00:10:47.509 Lisa Temple: frankly, just liked them, and liked what they were doing, and we collaborate with them, so they help us. It’s more of a friendship, but it’s also, like, been mutually beneficial, because they try new stuff out on us, and then we…
88 00:10:49.190 ⇒ 00:10:55.339 Lisa Temple: you know, bounce things off of them, so it’s been really great. It’s been a really nice relationship.
89 00:10:55.710 ⇒ 00:10:56.110 Robert Tseng: Cool.
90 00:10:56.400 ⇒ 00:11:04.630 Lisa Temple: So they’ve helped us a ton, thinking about how to set up.
91 00:11:04.810 ⇒ 00:11:14.340 Lisa Temple: our… our flat fees, we do use all fee. It basically… it’s pretty simple. I really want them to have more,
92 00:11:14.650 ⇒ 00:11:18.090 Lisa Temple: AI in there, but they do, it’s just not…
93 00:11:18.390 ⇒ 00:11:22.259 Lisa Temple: It’s hard to know, because flat fees are not,
94 00:11:23.530 ⇒ 00:11:30.820 Lisa Temple: you know, well, promoted by law firms, so I don’t really know what other law firms are doing, but also…
95 00:11:31.160 ⇒ 00:11:38.150 Lisa Temple: I know that the billable hour is going up an insane amount if you read anything about law firms.
96 00:11:38.390 ⇒ 00:11:41.220 Lisa Temple: And that’s just not the route…
97 00:11:42.270 ⇒ 00:11:46.360 Lisa Temple: I want to go. I don’t think the billable… I don’t like it
98 00:11:47.080 ⇒ 00:11:52.160 Lisa Temple: as an incentive. It’s incentive for bad behavior by lawyers.
99 00:11:52.450 ⇒ 00:11:53.200 Robert Tseng: And…
100 00:11:53.200 ⇒ 00:12:09.540 Lisa Temple: by… and it doesn’t make for good relationships, and so it’s always felt better to me than the billable hour. I’ve realized, however, it is super hard, to your point, because you asked, like, what are the things you may not be able to do by flat fee?
101 00:12:09.540 ⇒ 00:12:10.810 Robert Tseng: Yeah. And…
102 00:12:12.450 ⇒ 00:12:15.330 Lisa Temple: I feel like when I talk to most lawyers, they…
103 00:12:16.000 ⇒ 00:12:28.110 Lisa Temple: it’s so much work that they’re… and it’s so unknown that I get it. Like, there’s a lot of stuff we do that… oh my god, it’s so much easier to just to do hourly.
104 00:12:28.110 ⇒ 00:12:28.710 Robert Tseng: Yeah.
105 00:12:30.080 ⇒ 00:12:34.009 Lisa Temple: So, I would say.
106 00:12:34.320 ⇒ 00:12:41.329 Lisa Temple: we were kind of methodical with, Alti’s help to get our arms around what are the…
107 00:12:43.200 ⇒ 00:12:55.139 Lisa Temple: frequently occurring contracts that we have. We’re mostly… we mostly draft and review and negotiate contracts. That’s, like, 95% of what we do. We also do some, general counsel work.
108 00:12:55.250 ⇒ 00:13:02.419 Lisa Temple: Which is… you know, it could be anything. HR work,
109 00:13:03.840 ⇒ 00:13:06.680 Lisa Temple: advice on pretty much anything you can imagine. We’re generalists.
110 00:13:06.680 ⇒ 00:13:07.250 Robert Tseng: Yeah.
111 00:13:07.830 ⇒ 00:13:09.210 Lisa Temple: So,
112 00:13:09.340 ⇒ 00:13:17.829 Lisa Temple: But also, like, that means the contracts that we do are all over the place. But there are, like, 5 types that we’ve… we got it down, and we know what.
113 00:13:17.830 ⇒ 00:13:18.200 Robert Tseng: Mmm.
114 00:13:18.200 ⇒ 00:13:30.259 Lisa Temple: these are, and they’re, like, if you want it faster, then it’s this cost. If you have this many variables, then it’s more, or whatever. Yeah. So those kind of, like.
115 00:13:31.730 ⇒ 00:13:32.480 Lisa Temple: Hmm.
116 00:13:32.640 ⇒ 00:13:35.560 Lisa Temple: What, value add?
117 00:13:36.200 ⇒ 00:13:39.140 Lisa Temple: Things to the price, but…
118 00:13:40.100 ⇒ 00:13:54.130 Lisa Temple: the other stuff, like, we do big transactions… not big, I don’t want to, like, sound like we’re doing these, like, Wall Street-type transactions, but we’re doing, like, M&A stuff, right? That’s not small. And can you do a flat fee for that?
119 00:13:54.310 ⇒ 00:13:57.760 Lisa Temple: I don’t know. So, my friends at Alfie would… they…
120 00:13:58.030 ⇒ 00:14:11.700 Lisa Temple: frankly, the two guys who run it are brothers, and their dad owns a pretty big law firm in Vancouver, so they… he has done it. Like, he’s done all the… he does what… kind of what we do, but on a,
121 00:14:12.140 ⇒ 00:14:17.759 Lisa Temple: A larger scale, and what they do is stage the fees.
122 00:14:18.130 ⇒ 00:14:22.190 Lisa Temple: So that, you know, we’re gonna do this much work, and we’ll charge you this.
123 00:14:22.500 ⇒ 00:14:29.420 Lisa Temple: Or you get a success fee at the end. They do a lot that way, so that it’s,
124 00:14:29.520 ⇒ 00:14:33.250 Lisa Temple: It’s really tied to outcome, which is nice.
125 00:14:33.250 ⇒ 00:14:33.840 Robert Tseng: Yeah.
126 00:14:34.420 ⇒ 00:14:40.360 Lisa Temple: we’re just not that sophisticated yet, what to say. And it’s still…
127 00:14:41.120 ⇒ 00:14:44.810 Lisa Temple: It’s so much easier not to do it, so… .
128 00:14:44.810 ⇒ 00:14:45.330 Robert Tseng: Yeah.
129 00:14:45.330 ⇒ 00:14:50.740 Lisa Temple: Embarrassingly, I don’t do as many flat fees in that area as I would like.
130 00:14:51.250 ⇒ 00:14:57.809 Robert Tseng: Sure. No, I think that’s… that’s great context. Yeah, I think, I mean, I’ll look into AltB. I’m curious, like, kind of.
131 00:14:58.010 ⇒ 00:15:14.720 Robert Tseng: I mean, sounds like you already… I mean, within your practice area, even though you’re generalists, you have, like, you know, you’re able to see some patterns, like, you say 5 types of contracts that you’re kind of, like, going over and over again. So, I guess if they’re integrated with Clio, maybe they’re seeing that these are the types of, like, kind of…
132 00:15:14.720 ⇒ 00:15:31.060 Robert Tseng: transactions that you’re working on, and so that they’re able to recommend, like, hey, you should consider moving these into fixed… fixed fees. Have you seen that impact? Like, it seems like you said that the growth trajectory has changed, like, in the past year, so…
133 00:15:31.060 ⇒ 00:15:40.999 Robert Tseng: Does that… are you doing higher volume of the same contracts, or does that free you up to go and, like, do more of that other ad hoc work that’s, like, not part of, like, the set 5 templates?
134 00:15:43.510 ⇒ 00:15:49.010 Lisa Temple: I don’t really fully know what caused the uptick of work, except that I was doing more…
135 00:15:49.390 ⇒ 00:15:51.010 Robert Tseng: Video…
136 00:15:51.440 ⇒ 00:15:56.340 Lisa Temple: We upped our game on, like, LinkedIn and stuff.
137 00:15:56.340 ⇒ 00:15:56.850 Robert Tseng: See?
138 00:15:56.940 ⇒ 00:15:59.539 Lisa Temple: And so, it’s hard to really know.
139 00:15:59.810 ⇒ 00:16:06.100 Lisa Temple: But what we did, which is in the same family as flat fees, is subscription pricing.
140 00:16:06.880 ⇒ 00:16:08.099 Lisa Temple: But it’s…
141 00:16:08.560 ⇒ 00:16:17.449 Lisa Temple: Our subscriptions now, although we’re hoping there will be change in the future, are based on hours, so it’s not that different than hourly.
142 00:16:17.450 ⇒ 00:16:18.689 Robert Tseng: It’s like a bucket of hours kind of thing.
143 00:16:18.690 ⇒ 00:16:24.350 Lisa Temple: Yeah. Okay. It’s basically a retainer. But, but that…
144 00:16:24.970 ⇒ 00:16:34.330 Lisa Temple: area of the business, which is the most profitable took off, which is awesome. I don’t know if there’s any correlation, unfortunately, to…
145 00:16:34.640 ⇒ 00:16:36.239 Lisa Temple: our flat fee.
146 00:16:36.550 ⇒ 00:16:39.830 Lisa Temple: dreams and plans. Fair.
147 00:16:41.750 ⇒ 00:16:48.960 Lisa Temple: So… but yeah, I mean, to the extent that things come in that aren’t… so those… that work is, like, fractional general counsel.
148 00:16:49.480 ⇒ 00:16:55.189 Lisa Temple: And we’re in their systems, we’re doing, we’re in Slack, we’re in…
149 00:16:55.350 ⇒ 00:16:57.589 Lisa Temple: I don’t know if you’re familiar with Ironclad.
150 00:16:57.590 ⇒ 00:16:58.770 Robert Tseng: Yeah.
151 00:16:58.780 ⇒ 00:17:06.779 Lisa Temple: Wrike, some of their, contract management systems were kind of function as an employee of their company.
152 00:17:07.140 ⇒ 00:17:14.010 Lisa Temple: And so… It kind of… you know.
153 00:17:14.170 ⇒ 00:17:19.909 Lisa Temple: I don’t know, I’d rather it not be by hours, because we are using AI, to…
154 00:17:20.060 ⇒ 00:17:23.010 Lisa Temple: Do some of our contract reviews.
155 00:17:24.349 ⇒ 00:17:28.470 Lisa Temple: But, for other things that come in, like a one-off
156 00:17:28.710 ⇒ 00:17:33.690 Lisa Temple: draft of a lease, or something like that. We’re… we’re doing a mostly flat fee.
157 00:17:33.980 ⇒ 00:17:34.620 Lisa Temple: Yeah.
158 00:17:34.920 ⇒ 00:17:35.720 Robert Tseng: Got it.
159 00:17:36.640 ⇒ 00:17:37.480 Robert Tseng: Okay.
160 00:17:37.770 ⇒ 00:17:41.169 Robert Tseng: Yeah, I think, like, that’s…
161 00:17:41.300 ⇒ 00:17:45.980 Robert Tseng: That’s helpful to see that, like,
162 00:17:46.970 ⇒ 00:18:05.099 Robert Tseng: Yeah, I mean, there’s… at least for the fractional general counsel work, it seems like that’s… that hasn’t been really… you use automation in the background, but it’s not really, like, it’s… it’s not like dynamic pricing or, like, some AI… like, AI kind of informed pricing or anything on that side yet, just because that work seems like it’s…
163 00:18:05.440 ⇒ 00:18:18.870 Robert Tseng: you’re integrated… you’re involved in a lot more systems, it’s not just, like, purely off of, like, like, in Clio, you’re kind of having to go into whatever systems that your clients have, and maybe that’s… it’s harder to kind of get the diff… the…
164 00:18:19.750 ⇒ 00:18:29.980 Robert Tseng: well, I don’t know if Cleo really… if you really have, like, a knowledge base where everything kind of comes together, so it seems like these are, like, very different work streams. Would that be a good way to characterize it.
165 00:18:29.980 ⇒ 00:18:41.479 Lisa Temple: We do, and that’s creating a ton of work, admin work, actually, because we’re going into their systems, we’re downloading contracts, we’re uploading them to Clio, we’re compiling, you know, it’s like…
166 00:18:41.710 ⇒ 00:18:42.550 Lisa Temple: Which is why.
167 00:18:42.550 ⇒ 00:18:46.019 Robert Tseng: So you try to move everything into Clio. Yeah. I see.
168 00:18:46.900 ⇒ 00:18:50.480 Lisa Temple: And hence, we’re about to hire a paralegal, so if you know anyone, let me know.
169 00:18:51.050 ⇒ 00:18:51.570 Robert Tseng: Okay.
170 00:18:52.090 ⇒ 00:18:56.880 Robert Tseng: Yeah, I mean, my classmates, I’m sure, some people will be looking for jobs, yeah.
171 00:18:57.300 ⇒ 00:18:57.840 Robert Tseng: Yeah.
172 00:18:58.510 ⇒ 00:19:06.900 Robert Tseng: Cool! I mean, thank you so much for, kind of, the, the, the updates there. I think, that’s, that’s, that’s really interesting, to hear how you, how you’re adopting it.
173 00:19:07.270 ⇒ 00:19:12.239 Lisa Temple: Yeah. Well, tell me about… it sounds… John, tell me a bit about what you’ve been doing.
174 00:19:12.240 ⇒ 00:19:12.820 Robert Tseng: Yeah.
175 00:19:13.090 ⇒ 00:19:17.310 Lisa Temple: That you, I guess… what,
176 00:19:19.510 ⇒ 00:19:28.689 Lisa Temple: the way I was understanding it is that you take data, and you kind of normalize it, and optimize it for AI. Do I have that right?
177 00:19:28.870 ⇒ 00:19:32.710 Robert Tseng: Yep, that’s a very good… yeah, that’s a good punchline, we should use that.
178 00:19:33.560 ⇒ 00:19:39.209 Robert Tseng: Yeah. Yeah, I would say, like, kind of the… what that really means is, like,
179 00:19:39.710 ⇒ 00:19:43.120 Robert Tseng: We… we start with, like, building, like,
180 00:19:43.220 ⇒ 00:19:58.580 Robert Tseng: like, there’s different ways to call it. People call it single source of truth, knowledge base, like, data clean rooms, like, depending on your industry, like, everyone calls it differently. I don’t actually really know what they call it in the legal world. But yeah, I think it’s just, like, you know.
181 00:19:58.630 ⇒ 00:20:15.310 Robert Tseng: you have a case management system, like a Clio, and you’re kind of… it’s become that, like… it’s become that place for you where you’re trying to upload everything into. For us, we’re, like, more tool agnostic about it. We work with, I mean, healthcare, CPG, software, like, we think that this, like.
182 00:20:15.310 ⇒ 00:20:23.110 Robert Tseng: From a first principles perspective, this kind of applies to every industry, where every organization just needs, like, an index
183 00:20:23.110 ⇒ 00:20:40.880 Robert Tseng: knowledge base that AI can actually go through in a reliable way. And so, yeah, I guess, like, to your point, maybe people were unsure, like, can I really just open ChatGPT, throw everything I want in there, and then get something reliable, and is it even safe? Like, is the data protected? So…
184 00:20:40.880 ⇒ 00:20:48.699 Robert Tseng: I think, like, we… I think we take it much, much further than that, where, yeah, it even starts from just, like.
185 00:20:48.920 ⇒ 00:20:54.909 Robert Tseng: from, like, a context management perspective, being a data person at heart, like, when, like.
186 00:20:55.150 ⇒ 00:21:12.500 Robert Tseng: when we get all these different, disparate data sources, like, how do you rank the priority of them? Like, where… where… where… where… like, where do you… where do you put, like, the best examples? Like, where do you put instructions, etc. Like, all… it’s all, like, kind of in the background, kind of building, like, your internal wiki, for your… Okay.
187 00:21:12.500 ⇒ 00:21:13.250 Lisa Temple: Yeah.
188 00:21:13.410 ⇒ 00:21:23.599 Robert Tseng: For your company, and then getting it to a place where it’s not… I guess before, it was, like, for people to go and, like, search it up themselves. It relies on search queries and things like that.
189 00:21:23.830 ⇒ 00:21:24.300 Lisa Temple: Yeah.
190 00:21:24.300 ⇒ 00:21:35.039 Robert Tseng: like, now it’s much easier to prompt. You can just use any sort of, like, agentic interface that connects to, like, some MCP server that can read your knowledge base very… much faster than a person can.
191 00:21:35.110 ⇒ 00:21:44.489 Robert Tseng: So, yeah, those are the types of solutions that we’re building for clients now, where, you know, I think, like, if you were to…
192 00:21:44.820 ⇒ 00:22:03.510 Robert Tseng: Yeah, let’s say, like, for your situation, you know, Wrike, Wrike, Slack, Clio, like, you need to grab data from one system, move it to another, and it needs to be, like, kind of reformatted in a way that it’s actually ingestible. So, like, that could all be automated pretty easily at this point. And so…
193 00:22:03.510 ⇒ 00:22:21.379 Robert Tseng: Yeah, I think we do a lot of that type of, like, workflow messaging automation to kind of help different systems that don’t usually talk to each other be able to, like, kind of actually integrate with one another. And yeah, so that’s, like, all on the back end. On the front-end side, like, we build on the application.
194 00:22:21.380 ⇒ 00:22:24.280 Robert Tseng: layer as well, so I think it’s… you were…
195 00:22:24.280 ⇒ 00:22:38.170 Robert Tseng: you know, I mean, I could show you maybe another time on, like, some of the stuff that we build for ourselves internally, but, I mean, we do a lot of, like, kind of, you know, online marketing as well, whether on LinkedIn or through email, and so…
196 00:22:38.170 ⇒ 00:22:53.339 Robert Tseng: Everything from producing, like, a campaign brief off of an idea that I have, it kind of searches through our knowledge base for all the transcripts when I’ve talked to things related to the idea that I have. It’ll put, like, a first draft of a campaign brief to my go-to-market lead.
197 00:22:53.340 ⇒ 00:23:11.119 Robert Tseng: Then he can kind of, like, go through that first pass, figure out, like, directionally if this is, like, what pieces he wants to keep, and then he needs to be able to pass it on to, like, our brand marketing team that’s able to create, like, the beautiful assets when we’re doing video or we’re creating, like, white papers or other images.
198 00:23:11.120 ⇒ 00:23:19.889 Robert Tseng: And then, like, a whole content series as well. And so, like, we have, like, multiple agents that are kind of, like, assisting at each of these… each of these steps.
199 00:23:19.930 ⇒ 00:23:28.640 Robert Tseng: So, you know, in short, at a previous company, that would have been, like, 5 different people’s jobs. Like, I just have to have one person who’s kind of, like, doing… who’s kind of…
200 00:23:28.640 ⇒ 00:23:30.130 Robert Tseng: More like the…
201 00:23:30.130 ⇒ 00:23:52.020 Robert Tseng: he’s kind of like the quarterback, kind of directing all of these different team members that we’ve assembled to kind of, to kind of execute on the task. So, now, like, I now see, like, the… like, it’s possible to have, like, multiple agents working simultaneously to actually replicate an entire workflow. So that’s what I’m most excited about, and, like, what I’m interested in the legal space.
202 00:23:52.080 ⇒ 00:24:08.330 Robert Tseng: Because, yeah, I think, like, contract drafting and, like, a lot of these, like, kind of edits on templates is, like, a good step one for what AI can do, but I’m really interested in getting into the weeds with somebody in, like, a end-to-end, like, more complex or bespoke workflow, and trying to see, like.
203 00:24:08.400 ⇒ 00:24:24.419 Robert Tseng: how could I break that up into a bunch of individual agents so that, like, you’re able to speed up that time by, like, you know, 80, 90%, and you just kind of are just a reviewer at that stage? So, you know, I think that’s… that’s kind of, like, my interest in, like, the…
204 00:24:24.420 ⇒ 00:24:29.239 Robert Tseng: what… what it would be like to build in the legal world. But… but yeah, I guess that’s kind.
205 00:24:29.240 ⇒ 00:24:32.599 Lisa Temple: That is very exciting. Yeah.
206 00:24:32.870 ⇒ 00:24:44.850 Lisa Temple: it sounds… are you in law school because you want to work in the… like, why did you go to law school? Do you want to practice? Like, you are so… you’re gonna make so much more money as, sticking with what you’re doing.
207 00:24:45.380 ⇒ 00:24:46.479 Lisa Temple: I do it!
208 00:24:46.480 ⇒ 00:25:03.810 Robert Tseng: Yeah, yeah, I don’t know if I want to be a practicing attorney. I thought, like, law school would give me more exposure to, like, learn, like, the… like, what do lawyers do, and, like, what practice area would this be really beneficial for? And I think in some… to some extent, it has been helpful, like.
209 00:25:03.810 ⇒ 00:25:09.809 Robert Tseng: I’m… I’m… yeah, like, in… in narrowing down, like, in this whole, like, in the legal world, like, what…
210 00:25:10.290 ⇒ 00:25:16.849 Robert Tseng: like, where would I like to work specifically? I don’t think I know, but I feel like…
211 00:25:16.990 ⇒ 00:25:41.970 Robert Tseng: I mean, building, like, pro se, like, AI tools is interesting for courts. Like, I’m actually doing something with the First Circuit right now to kind of help them, like, it’s, like, more, like, for, tenant, tenant-landlord relations to fill out a very specific type of form and getting them to… for self-represented litigants. So, like, that’s… that to me is an easy project. It’s just helping people fill out forms better and, like, directing them what to do in the
212 00:25:41.970 ⇒ 00:25:42.630 Robert Tseng: complex thing.
213 00:25:42.630 ⇒ 00:25:43.260 Lisa Temple: Yeah.
214 00:25:43.260 ⇒ 00:25:51.540 Robert Tseng: But, like, yeah, like, I still feel like that feels very, like, entry-level, and, like, I would prefer… I mean, I’d be interested in
215 00:25:51.540 ⇒ 00:26:09.429 Robert Tseng: digging into something that is, like, a really complicated, really fast-moving, like, subject or domain area, like immigration, or like, I guess people say tax law changes all the time. So, I feel like that’s where the interesting… I mean, I feel like that would be more interesting to me, but I really don’t know yet.
216 00:26:10.740 ⇒ 00:26:20.880 Lisa Temple: Wow, I love it. You’re thinking on, like, so many different levels, and, you’ve followed your curiosity, which is so awesome. I have,
217 00:26:21.760 ⇒ 00:26:23.549 Lisa Temple: concept… I used to work…
218 00:26:24.220 ⇒ 00:26:33.909 Lisa Temple: I’m trying to think of, as you’re talking, like, how can I help you? How can I… Sure. I’d love to work with you at some point, I don’t know what that means, or, like, I don’t know, I just want to be of assistance.
219 00:26:34.310 ⇒ 00:26:35.820 Lisa Temple: So,
220 00:26:36.580 ⇒ 00:26:44.689 Lisa Temple: I do… I used to work at Thomson Reuters at Practical Law. Do you know what that is? So I… it’s like the,
221 00:26:46.650 ⇒ 00:26:58.890 Lisa Temple: I felt very new and cutting edge at the time, but it’s old now. But their whole thing is, like, be practical about how you practice law, and they were acquired by Thomson Reuters, but they were a startup at the time when I worked with them.
222 00:26:58.890 ⇒ 00:26:59.450 Robert Tseng: Huh.
223 00:26:59.680 ⇒ 00:27:04.790 Lisa Temple: So, all of the people who work there are super into legal tech.
224 00:27:04.790 ⇒ 00:27:05.590 Robert Tseng: Yeah.
225 00:27:05.590 ⇒ 00:27:15.570 Lisa Temple: And then, I have a lot of clients in the health and health tech area, that they’re just starting out, and they’re brilliant, and they…
226 00:27:16.060 ⇒ 00:27:20.600 Lisa Temple: are doing really cool stuff to make the world better. Like… Yeah.
227 00:27:21.810 ⇒ 00:27:29.130 Lisa Temple: Delivering… using software to help clinics deliver care in a more holistic way, or…
228 00:27:29.130 ⇒ 00:27:29.460 Robert Tseng: Yeah.
229 00:27:30.840 ⇒ 00:27:39.750 Lisa Temple: helping employees find the right healthcare benefits, you know, through, you know, developing software that way. Like, just…
230 00:27:40.420 ⇒ 00:27:45.330 Lisa Temple: kind of audacious, and…
231 00:27:45.700 ⇒ 00:27:54.640 Lisa Temple: interesting things, which I… I really love working with people who do that kind of thing. So it’s inspiring to me, even though I’m the one who’s like.
232 00:27:54.640 ⇒ 00:28:06.479 Lisa Temple: doing the boring, toiling work of contract review, but at least I’m working with people who are doing some awesome things, and it sounds like you’re… you’re in the category of, like, doing awesome things. That’s awesome.
233 00:28:07.720 ⇒ 00:28:08.300 Robert Tseng: Thank you.
234 00:28:08.300 ⇒ 00:28:13.980 Lisa Temple: to the extent I can help you and connect you with someone, I mean, I don’t know, my LinkedIn network.
235 00:28:14.880 ⇒ 00:28:22.180 Lisa Temple: is, you know, available. If you want me to connect you with anyone that you see who might be interesting to you, let me know, or…
236 00:28:22.520 ⇒ 00:28:25.870 Lisa Temple: I’ll stop talking, and you can tell me what your thoughts are.
237 00:28:26.150 ⇒ 00:28:40.660 Robert Tseng: Yeah, yeah, no, I think that would be helpful. If anyone comes to mind specifically, they’re like, hey, former colleague, legal tech, like, talk to them. I’d love to just even have an open-ended conversation with them. Yeah, maybe I’ll tell you, like, specifically areas… I mean, health and health tech is something…
238 00:28:40.660 ⇒ 00:28:48.760 Robert Tseng: I mean, half our business is in health right now, so, like, I feel like I have some domain expertise there that I could follow that and go a little deeper that way.
239 00:28:48.780 ⇒ 00:29:06.700 Robert Tseng: Yeah, and then I guess, like, for you, you know, in your firm, like, you know, as you’re kind of thinking through, I can tee it up with, like, some of the questions, like, hey, if you’re running into some of these questions, like, maybe it’d be a better time for us to chat. Sounds like you’re pretty bought into the Clio ecosystem, and I’m not, like, trying to…
240 00:29:07.070 ⇒ 00:29:18.760 Robert Tseng: I mean, it’d be interesting to, like, try to kind of see what it’s like to work alongside that, but, yeah, it seems like you have, like, a lot of process around… around that already, but if there’s any kind of, like.
241 00:29:18.760 ⇒ 00:29:30.889 Robert Tseng: bespoke work or growth opportunity that you’re kind of like, hey, new service line I want to test, like, how do I bring AI into something that, like, you feel like is not working yet? I’d be… I’d be interested in being a thought partner there and seeing, like.
242 00:29:30.890 ⇒ 00:29:45.050 Robert Tseng: at least telling you, like, I think this is how we could approach it, and see if that, like, is an interesting, you know, bet that you want to make. So, I think that would be, you know, a way that we could continue our conversation, too.
243 00:29:45.470 ⇒ 00:29:52.040 Lisa Temple: Awesome! I love that. And, yeah, I have a lot of pain points right now that I’m trying to iron out with my…
244 00:29:52.270 ⇒ 00:29:57.629 Lisa Temple: With my firm, my business, the business of the firm. And,
245 00:29:58.310 ⇒ 00:30:04.920 Lisa Temple: you’ve sparked a couple of thoughts, in this conversation that I might follow back up on. I just…
246 00:30:06.260 ⇒ 00:30:09.009 Lisa Temple: Too early days of these thoughts.
247 00:30:09.010 ⇒ 00:30:11.510 Robert Tseng: Yeah.
248 00:30:11.730 ⇒ 00:30:16.700 Lisa Temple: But that’s great. Are we… are we connected yet on LinkedIn?
249 00:30:17.430 ⇒ 00:30:20.479 Robert Tseng: I don’t think so yet.
250 00:30:21.050 ⇒ 00:30:24.220 Lisa Temple: Okay, well… Let’s do that.
251 00:30:24.530 ⇒ 00:30:25.190 Robert Tseng: Yeah.
252 00:30:25.440 ⇒ 00:30:26.619 Robert Tseng: I’ll find you.
253 00:30:27.190 ⇒ 00:30:29.980 Lisa Temple: Okay, I would love to, and we’ll just…
254 00:30:30.870 ⇒ 00:30:32.720 Lisa Temple: See how we can help each other.
255 00:30:34.330 ⇒ 00:30:36.909 Robert Tseng: Okay, I just sent you a Kinect.
256 00:30:38.500 ⇒ 00:30:39.650 Lisa Temple: Terrific.
257 00:30:39.650 ⇒ 00:30:41.150 Robert Tseng: Yeah.
258 00:30:43.550 ⇒ 00:30:51.350 Robert Tseng: follow you along. Yeah, congrats on, like, kind of all the, yeah, seems like your firm’s on to exciting things,
259 00:30:51.630 ⇒ 00:31:05.170 Robert Tseng: Yeah, no, I’m very keen on following, kind of, your journey and seeing what works, doesn’t work for you, so I’d love to stay in touch. I’m assuming you’re based in Connecticut. I don’t know how often you make it into the city, but.
260 00:31:05.170 ⇒ 00:31:06.050 Lisa Temple: I do.
261 00:31:06.050 ⇒ 00:31:08.169 Robert Tseng: Whatever of time, we’d love to… love to meet.
262 00:31:08.530 ⇒ 00:31:20.259 Lisa Temple: Awesome! That’s great. We… John comes in more than I do, but I do… most of my clients are in New York City, so, I have to bite the bullet, and I use the snow as an excuse, but it’s gonna melt soon, so I have to…
263 00:31:20.630 ⇒ 00:31:34.050 Lisa Temple: I hate going into the city, but I will do it. I need to get out there and get out of my house, which is good all around. So, I’m sure John would be in agreement, so,
264 00:31:34.660 ⇒ 00:31:36.529 Lisa Temple: Good idea.
265 00:31:37.700 ⇒ 00:31:49.390 Robert Tseng: Okay, yeah, let me know, I’d love to take, you know, take you guys out for a meal, or just kind of get to know, get to know you a bit more, and person here. But yeah, otherwise, like, yeah, I look forward to staying in touch.
266 00:31:50.050 ⇒ 00:31:51.650 Lisa Temple: Great, Robert, good luck.
267 00:31:52.160 ⇒ 00:31:53.030 Robert Tseng: Thank you.
268 00:31:53.030 ⇒ 00:31:53.360 Lisa Temple: Okay.
269 00:31:53.360 ⇒ 00:31:53.760 Robert Tseng: Obviously.
270 00:31:53.760 ⇒ 00:31:54.600 Lisa Temple: Yeah, bye.