The MAD Podcast: How AI Gets Built — with Matt Turck
00:10:27 · George Sivulka
[00:10:27] S1: Yeah, so, know, DefM14A wasn't the end all be all from those earliest days. And honestly, we spent a very long time trying to find the perfect market niche, you know, our equivalent to fixed income, where Excel in like the span of eighteen months took over 90% of the fixed income industry and like everyone was using it and they started using it for everything else. And so it was like, what... Where is it? Where where could this technology be immediately leveraged where we could go to market, where we could take over 90% of the market as fast as possible and then go and be leveraged for other things? And the way that, you know, our team really thought about it was, hey, these large language models and the tool that we've created is very good at mapping from one ontology to another ontology. You know, one set of unstructured data, some taxonomy, some set of files into, hey, here's a list of questions and a list of answers. And in private equity due diligence, have a data room, which has all of the company's files, and it's your job as a private equity analyst or a third party to come in and fill out a 100 or 400 question long tracker that has all these questions about a business. So in private equity, there's no numbers or reporting or anything else, it's all in the structured data and you have the people that are the highest paid that hate their lives the most, whose jobs is entirely to, hey, go map from this data room about a company to, you know, this set of criteria that dictates whether or not we pour hundreds of millions of dollars into it. And so we started there. And so, we've kind of gone to market with many of the largest asset managers in North America and now Europe, and we solve basically their most pressing pain. How do you deploy more capital more quickly in a more accurate fashion? And six months ago there wasn't a single private equity CEO who'd, you
00:12:28 · Matt Turck, George Sivulka
[00:12:28] S2: know, even [00:12:28] S1: talk about heavy and now they're all like, I need a large language model strategy and I've heard that my team is getting beat by this other team that has it, and it's helping us go to market really, really fast, which is quite good. [00:12:40] S2: Are you finding that the best deployment method for this is a Copilot kind of analogy or are you actually enabling people to just replace those unhappy analysts?
00:12:40 · Matt Turck, George Sivulka
[00:12:40] S2: Are you finding that the best deployment method for this is a Copilot kind of analogy or are you actually enabling people to just replace those unhappy analysts? [00:12:57] S1: So I think the thing that I like fundamentally very much believe, which may be the kind of the last demo, you know, made me believe a little bit less, but yeah, is that humans are really, really important and making humans better is actually the ultimate goal of what I do. And if you look at, you know, I'm going to harken back to Excel once again, bookkeeping before Excel and the structured database was literally in books. You know, a spreadsheet is called that because the spreadsheet was spanned over the two different pages, and so you'd open up a big book and you'd see a spreadsheet. And when this technology came out and when it was easily available in these productivity tools, people's jobs changed, and their job actually became, hey, I'm not going to use a slide rule and all kinds of calculators to do my job, instead I'll use a computer. And I think Hebia is very similar. Like, we are building a tool that is aligned with humans and aligned with creating value. And I think, to date, not a single job has been replaced by Hebia, knock on wood. But I think one day there will be many more jobs that are created to run Hebia.
00:12:57 · George Sivulka, Matt Turck
[00:12:57] S1: So I think the thing that I like fundamentally very much believe, which may be the kind of the last demo, you know, made me believe a little bit less, but yeah, is that humans are really, really important and making humans better is actually the ultimate goal of what I do. And if you look at, you know, I'm going to harken back to Excel once again, bookkeeping before Excel and the structured database was literally in books. You know, a spreadsheet is called that because the spreadsheet was spanned over the two different pages, and so you'd open up a big book and you'd see a spreadsheet. And when this technology came out and when it was easily available in these productivity tools, people's jobs changed, and their job actually became, hey, I'm not going to use a slide rule and all kinds of calculators to do my job, instead I'll use a computer. And I think Hebia is very similar. Like, we are building a tool that is aligned with humans and aligned with creating value. And I think, to date, not a single job has been replaced by Hebia, knock on wood. But I think one day there will be many more jobs that are created to run Hebia. [00:14:20] S2: Sure. You alluded to some of this already, look, I don't know how much you can sort of disclose, but do you want to talk a little bit about what's under the hood in terms of what you built, how it works, what was hard, what turned out to be easy, like anything you can share on a more technical level. [00:14:41] S1: Yeah, I'm happy to take questions as well. I don't know if that's something we'd...
00:14:41 · George Sivulka, Matt Turck
[00:14:41] S1: Yeah, I'm happy to take questions as well. I don't know if that's something we'd... [00:14:45] S2: Yeah, we can do this afterwards. Like maybe start and then go.
00:14:49 · George Sivulka
[00:14:49] S1: The three buckets that I can definitely talk to are input, output, and then flexibility. And these are all kind of different things we've done a lot of R and D into. One thing I am obsessed about is not being replaced by an increasingly large language model, or better language model. Kind of side anecdote today, OpenAI released, as many people thought they would, a new API that allowed you to call functions, right? And so it was kind of a big thing and there's a ton of startups that had done a lot of work into doing R and D around how to create a tool form or implementation, yada yada yada, and they released this today and that works on mute. So, what I care a lot about is how do I build tooling and functionality around a large language model to make it much better? On the input side, just like there's entire companies, industries even built around getting data into a SQL database, We've done a lot of work on integrations, on parsing documents, on actually creating multiple types of indexes to make sure that these models have access to good information in a way that actually matters. On the output side, we've done a lot of work to find ways to actually successfully programmatically wrangle them to get to outputs that our users want, and to do that in a generalizable way where the user is still in control of what that output looks like. And then on the flexibility side, it's a beautiful product, in my opinion, that allows people to very easily understand what they're doing, right? And there's a lot of work into building reliability and trust for end users because people just don't, by default, trust the output of a large language model. And if their job is on the line, if they're allocating a ton of money or they're making a mission critical decision, it matters that they're right.