Episode 75: Raj Bakhru of BlueFlame AI on
How AI Is Transforming Private Equity Workflows

Episode Transcript  

Shiv: All right, Raj, welcome to the show. How's it going?

Raj: Good, thanks for having me, Shiv. 

Shiv: Yeah, excited to have you on. Obviously, AI is a very hot button topic right now in private equity. So why don't we start with an intro about yourself and BlueFlame and let's go from there.

Raj: Yeah, sure. So my personal background is actually on the tech side of things. I started my career at Goldman in the quant hedge funds, building FX trading models and algos and research systems and things like that. Back when we were one of the world's largest quant funds. Went to Highbridge Capital after that, and then Capos Capital, which was a spin out Goldman fund building research and trading systems there. Started a security firm in 2014 actually that was bought by ACA compliance group in 2015. And you probably don't know ACA, most people on this, that listen to this probably don't know ACA, but it's the largest governance risk compliance provider to the buy side entirely and sort of started in the alternative space with private equity and hedge funds. Built out a bunch of stuff there, built out all of our privacy, reg tech offerings, our ESG business line, and then eventually ran our M&A and corporate strategy under our private equity owners. had New Mountain as our first private equity owner. And then they sold to CV Star who sold to Genstar. And I was running our Corp Dev and M&A and strategy. So we got to see a lot of the fun private side of the world. ChatGPT hit, a bunch of us looked around said, this is going to completely transform what's going to transpire and how work gets done in private markets. So we felt like we had really good pedigree to get after it between our quant backgrounds as well as our compliance risk privacy backgrounds. It's a big headwind on AI adoption. So we started building BlueFlame about two years ago and been out in market for about a year and a half now. BlueFlame's at this point working with about 75 alternatives managers, mostly in private markets. We're about 40 people helping execute on use cases that are really specific to this space that we can talk about.

Shiv: Yeah, and for what it's worth, we normally don't have platforms on the podcast. We have either private equity operating partners or GPs or maybe service providers to this space. So walk us through BlueFlame in particular. And the reason we've had you on here is just because the AI component and obviously you're working with private equity. So just help us understand what does the product do and how does it add value to the ecosystem here?

Raj: Yeah, sure. So we help with agentic automations in private equity. And there's a number of things that you need to execute on that promise of agents are going to help us build out workflows and automate parts of our lives that couldn't be automated before. The technology, the LLM technology lends itself really well to the space because of the massive unstructured data that we all sit on. So every data room that we all have, you typically have a folder portfolio company, a folder per fund, folder per LP, just tons and tons of PDFs in Excel and Word documents that are across all these folders. And now we have the ability to harness all of that data, that treasure trove of data that people have sitting on their drives and in their CRMs and so on. So a lot of what we've done with BlueFlame is build a connectivity layer to all of these data sets, power them with LLMs to execute use cases that matter to our client base. We're a little bit unique in that we're servicing every persona within the private equity firm. So we tend to start with the investment teams and research and diligence and help with sourcing. We'll help with the fundraising and IR teams with a bunch of their workflows that we could talk about. But we actually go all the way to back office. We're providing basically LLM power to everyone in the firm that's really specific and understands our space.

Shiv: And how does that work? Because there's a ton of different data sets, right? So what are you kind of plugging into and what kinds of insights are you coming back?

Raj: Yeah, so we think about the data element, sort of the unsexy piece of this, which is critical as five buckets. You have your unstructured file system, so that's your cloud storage or box, Dropbox, SharePoint, Ignite. We connect into all of those. We sync those in. We bring permissions in on those, which is pretty unique. Our clients often have terabytes of data sitting on their drives that they've never really been able to ask questions against. So what we do with that data when we bring it in is we'll parse out all of the really interesting content, we’ll look at the charts, the graphs, the logos will pull it, pull those into vision models to understand them. We pull out tables and we take all the text. We enable question answering against all of that text. So that's the first data source. Second data source is public data that's out there. Public market data that's out there. So think earnings transcripts, 10-Ks and Qs often used for things like valuation and comps. The third type of data are third party market data sources. So think Grata, PitchBook, Prequin, CapIQ, we have the ability to pull those in. And then fourth is what we think of generally is the books of record. And for most people, the biggest one there, the biggest two there are Office365, email, calendar, obviously tons of really interesting information in there, as well as your CRM. And some of our clients actually have multiple CRMs. have their investor CRM and they have their deal CRM. We're usually connecting into both of those. Sometimes, you know, you have a book of record that's like your portfolio accounting or monitoring tool, like a Chronograph or Cobalt. We connect into those. And then the last element is all of the public sources of the LLMs, Google, Google News, scrape LinkedIn profiles. And really our thesis and what we found to be true is that if you were to look across these five buckets of data, you can execute 90% plus of the workflows in our space. Right, when you pair them with reasoning models and agentic reasoning and thinking.

Shiv: How do you see that coming into the workflow? And it's more of general question about AI as well as there's just so much context that sits inside data rooms, inside PitchBook and the analysis that humans, I guess, or experts are doing on a particular business. How do you replace that with LLMs and more generative AI tools to be able to do that level of diligence to be able to really understand, let's say the enterprise value of a company, quality of earnings or whatever else that you're trying to analyze.

Raj: Yeah, and we're not going to claim that we can replace those bodies of work yet. And in particular, there's where that reasoning is really necessary around things like I've seen deals like this in the past. These are the red flags that we've seen on these deals in the past. We're still a ways away from being able to replace any of that type of judgment. But what we can do is supplement it, right? We can bring in automation to a lot of the things that are frankly just regurgitation of information. If you think about scorecards, IC memos, one-pagers. It's often, let me go to the data room, aggregate a bunch of the data, supplement it with things out of maybe my DealCloud or Salesforce, supplement it with things out of Browder or PitchBook, put that together in a slide or two, three, four pages of slides that have a specific format that we follow. And then I can supplement that with views on, this is what I think the upside case is, this is what I think the downside case is, this is why we might be concerned, these are the areas that we need to diligence further. That human element is still very valuable today that the AI technology is not replacing that yet. Over time, it'll obviously get smarter and understand more of that. And we could talk about the path to doing that with agents, but that's not where it is just yet.

Shiv: Got it. Okay. And just, guess that in that case, it's almost like step one of the analysis process, but then it's kind of handed back into the workflow of the private equity firm. But that's step one you're saying kind of takes time for a PE firm to kind of put together as an analyst is sifting through a data room to analyze things like.

Raj: Yep, we had a partner at one of our private credit clients this week who said the IC memo generation that we did for him saved him 10 hours of work on the weekend and he got to go play tennis because he didn't have to go do that 10 hours of data aggregation. We didn't go and put in the thesis, the investment thesis. We put in all the other data to support that investment thesis.

Shiv: Got it. Yeah. Yeah. And just zooming out, just away from this BlueFlame and just thinking about private equity in general. And I wanted to go into your business in particular, just to kind of set the stage, but for the audience as they're thinking about their respective firms and thinking about AI in general, how do you see them bringing in AI tools into their workflows? Because, you know, we, for example, we are also producing our own AI tool. It's called Inquisio and it allows PE firms to do marketing due diligence on potential targets that they're looking at. And we built a massive database and almost like an agentic application that gives them an idea what the right size budget and strategy should be for a particular company. But I see this as a pattern that should exist across all PE firm activities as well. So that's kind of why we made the bet. But I'm curious, like on your side, how do you see that evolving and how do you see the PE firm in general evolving in a world where more AI agents and these types of tools are emerging?

Raj: Yeah, and I think this has changed over time. People have gotten much more mature and they're thinking about how AI is going to affect their firm. You know, routinely we're seeing it as a topic at offsites where they're working through their own AI roadmaps. Almost every firm we're working with has an AI committee. At this point, that AI committee has figured out these are our top use cases. We've stack ranked them in terms of ROI to us. Almost always because the investment team is so expensive and their time is so valuable, the investment use cases go to the top, especially the ones where, you AI is already in a place where it can service that. And what they've learned over time, and we've all collectively learned over time, is that the technology in some ways is magical, right? It's incredible what it can do. In other ways, it takes a lot of iteration. It's a true technology project to adopt and change your workflows to leverage AI. So you can't just sort of implement something, have it be perfectly turned key and change everything about your firm tomorrow. That's just not gonna happen. People recognize it's gonna take back and forth iteration, training, guidance, prompting, teaching people how to use a new technology, iterating with the outputs, things like that to get to exactly what you want. Which also means that you can't change the entire firm overnight. You have to pick off your top three to five and focus on those three to five. Now for some firms that's, you know, diligence items, many firms that's diligence items, right? So helping with marketing diligence or really any of the other work streams on diligence. For a lot of firms that's, you know, our strategic priority for this year is sourcing more deals in a prop manner, right? Just getting more prop deal flow. And how you get more prop deal flow with AI is leveraging it to do more of your outbound outreach, as well as more identification of targets. So you focus on those use cases and the data and the connectivity required to execute those use cases.

Shiv: Yeah. And so what are those top areas that you're seeing PE firms kind of really zero in on to leverage AI where they think can add the most value. I asked because I'm sure there are patterns that the audience can kind of take away from it say, maybe I should be focusing on some of these things to create more value for the firm.

Raj: Yeah, and what I will say is it varies from the firm based off their own workflows and how they do things. So what you'll actually find, you know, just as a simple example, some firms do tons and tons of expert network calls, and they're willing to take in those transcripts and they want to leverage the history of those transcripts to ask research type questions against. Whereas other firms have a policy of zero expert network calls or have a policy if they're not allowed to have the transcripts, in which case obviously that use case is less meaningful to them. So the areas where we tend to see a lot of excitement and success, you know, sourcing, as I mentioned, one of those. What a lot of people are doing with sourcing is actually, you know, end to end. So it's taking the banker deck, automatically parse out the names in that banker deck, or taking the industry conference attendee list, parse out all the companies on that attendee list, have it go to my Salesforce or Deal Cloud CRM, auto add entries into the CRM, have it figure out who the management team is of those firms, find a description of those companies, draft a hyper-personalized email out to the CEOs of those companies, if we haven't had contact with them already. You can actually execute most of that workflow today with where the AI solutions are, and that's super valuable to a bunch of our clients. We've seen, you know, interesting examples where some firms, because they put so much emphasis on sourcing, they've historically said, you know, managing directors get a better response rate on sourcing emails than, you know, an associate or an analyst will. So we're actually going to have the associate or analyst sit down at the managing director's desk and draft each one of those emails out to those target CEOs so that we get more responses. And now that entire process is obviously, you know, much more, much closer to automated. They still review the emails before they go out, but you can generate basically the exact same emails as were being generated before. So that one's super powerful. That one's very common. 

Deal teams almost uniformly have some form of output. Initial output office CIM or initial data room, whether that's a one pager deal memo and word. Maybe that's a deal, what's called a heads up memo at some firms. Maybe it's a scorecard. Maybe it's a one pager. Maybe it's your initial IC deck. Whatever it is that tends to live in either Word or PowerPoint tends to be one to five pages of content. That's largely a regurgitation of who is this firm? What do they do? What are their products? Who's the management team? What are the financials? What's the outlook? And all of that type of information can be ascertained automatically by the AI and put into that format for you. Now that takes work to iterate to get to exactly the tone and the format and the style that the firm wants, but that's a big focus area for a lot of firms. 

Search is probably actually the easiest one for everyone to understand. What we think of as global search is a use case in its own right. A lot of firms just don't know where the data lives. They know I've seen a deal in this space before. Don't recall what it was. Don't recall any details about it. How can I just search for all the right tech deals we've seen in the past? So we have, have, you know, basically the ability to go across those five buckets that I talked about. can find those deals really anywhere. Research is obviously the next level of search. And we think about research as agentic search. So what you do with research is you basically take a topic, decompose it into areas that you want to understand, run searches against each of those areas, compile a report about all of that. A lot of people have seen deep research. It's basically deep research, but on your internal content. We've seen people who built just monitoring of news, earnings transcripts, portco news, portco competitors, other GP competitors. That's all been pretty successful for DL teams as well.

Shiv: What about on the, on the ops side? I totally get the deal sourcing and some of the front end work that is being done inside private equity firms. But it's almost like this, that's like the first step, right? Similar to how we were talking about initially analyzing data rooms, but there's all this other work that happens throughout the value chain inside of a private equity firm. And one of the biggest areas is value creation. So talk about that. Like where are you seeing some of the biggest use cases for AI on that, on the value creation side?

Raj: So a lot of the operating teams are leveraging it against board decks, against other content coming out of the portfolio companies to understand the status of initiatives, tracking of those initiatives and comparisons across other portfolio companies or deals they've seen in the past. So if you think about a firm that's struggling with sales in a particular geography and you know that you've seen other deals that have sold into that geography or you've had other portfolio companies selling into that geography, you can now find those examples pretty readily. You can dive into those examples and understand them much more rapidly than you could in the past. So I think that's one big focus area. The second is just reminding yourself, frankly, as a board member of these companies, of what are the things I should be thinking about? Where should I be challenging them? What are the questions I should be asking? So you often find cases where you're going into the board meeting and you're trying to remember all of the initiatives from a quarter ago's board meeting. And you want to hold them accountable to many of those items. It's actually now just one sentence to say, what were the initiatives from the prior board meeting that I ought to be checking on at this board meeting? So you just can get a lot more clarity and insight into the stuff that you should be tracking as an operating team member. Meeting prep is a very common one. Support for add-on deals, add-on diligence, all of the same workflows, very common as well. And then as you dive into the different departments of the firm, so if you think about what should that firm be doing from a sales perspective or a marketing perspective, you can leverage a lot of the technology to understand what parallels exist out there.

Shiv: Right. Yeah, I think that is helpful. I think, you know, one thing that we see that as we run our firm and we work with private equity investors is we're getting pinged from a lot of firms that have a performing business that has seen a significant drop off in their pipeline because let's say paid search or organic social or organic search, excuse me, is driving way less traffic as people are using LLMs and ChatGPT and other platforms to answer questions that normally would have led to some sort of an SEO click to their website. And so we are trying to figure out how they can recover lost traffic for them and recover some of that loss pipeline. So that's an example of a challenge that we've seen as a pattern across many companies. I'm wondering, like, I'm sure you've come across that, but what are some other big AI induced problems that you're encountering and how that kind of affects how operating partners and PE firms are working to kind of recover some of the enterprise value or productivity lost inside these companies as well.

Raj: Yeah, I mean, that's definitely one that comes to mind that everyone's struggling with and trying to figure out now. I think, you know, as everyone has learned or as a vault has evolved over time, the LMS are now web integrated. So the running searches on their own within the ChatGPT interface or Perplexity interface or what have you. So you don't get those clicks, but what you do have the ability to get after are putting out blog posts or other thought leadership that gets picked up by the LLM content when it runs those searches. So I think that's how most people are addressing that headwind at the moment. And obviously you probably have deeper knowledge on what other techniques exist out there on that front. I think where everyone else is struggling is how do we AI enable our portfolio companies? Now, we don't help with that. We're strictly focused on the GPS themselves, but basically everyone has this opportunity and this monstrous risk that exists with every portfolio company right now of, are they going to be disrupted by AI or are they going to be able to leverage AI and dramatically improve margins and processes and workflows. So I think a lot of people are frankly spending a lot of time trying to understand that.

Shiv: Where are some of the areas where you kind of, and I'm kind of going through all these areas because as you're building your platform, I'm assuming that it's also giving you exposure to other platforms and more AI related problems in this space. Where are you seeing AI either reduce productivity or kind of make it more challenging for companies to grow? Or what are some areas where transformation is required to capture the value that sits inside these companies? And what are some things that PE firms should be considering. And I'll give you just an example to kind of respond to. I was recently at an event speaking and one of the CEOs asked me that, you know, they're like, well, I can see that my team size or the number of people I have on my marketing team actually needs to shrink. And we need to think, to think about content a lot differently. So how would you go about building a content strategy in an AI first world? So I'm trying to understand from your perspective, like what are you seeing in terms of the different types of transformations that operating teams should be prioritizing inside their companies as a result of some of this stuff.

Raj: So I think very few people have said, we've adopted AI, it's completely transformed what we've done and been able to change on a dime, if you will. And even the folks who claim they did that have walked it back, right? There have been firms who went out and said, yeah, we dropped 700 customer service reps because AI's handling everything. And then that's gotten walked back, know, months later where they've said, you know, yes, that's great, but maybe 700 wasn't the right number. It should have been, you 300. So I think people have now learned that it takes iteration, it takes a lot of customization. As magical as this technology is, there's a lot of cases where you find yourself banging your head against a wall too. So you have to work through all of those corner case issues that will emerge as you adopt the technology. And that takes time, it takes effort. So you're not gonna be able to, in one budget cycle, completely transform a department. It's just not gonna happen. But you do need to be investing into it for that transformation because you're also not going to be able to pick this up in three years and catch up to what everyone else has done. So you can't fall that far behind. You have to be investing behind it. Just recognize that the gains of that investment are going to be very big. They're just not going to be overnight.

Shiv: Do you anticipate these companies - there's multiple forces happening at the same time, right? Because you have, on the revenue side, you might have, let's say, lesser pipeline because these agents are solving some problems up front that users may have previously chosen a SaaS solution for, as an example. But then on the productivity side, you might need fewer people to run the business. So in terms of the transformation, like, net-net, do you think that the margin expands over time or does it contract for these companies based on all these different forces that are pulling the company in different directions?

Raj: I think necessarily it has to expand. People won't adopt it if it doesn't, right? It's time investment and the outcome of that needs to be margin expansion. But that again will take some time. I don't think it's a matter of we're dropping our entire team. I think it's a matter of refocusing that team. If you were to ask any managing partner at a private equity firm, you know, what's on your to-do list? What are your priorities? What do you want to get done? It's a very, very long list of high ROI items, and they're just not getting to all of those items today. They don't have the staffing to do it. They don't have enough bandwidth internally to do it. So what you're going to see happen is a number of those folks who are spending time on things that AI is going to take over are now going to redirect their efforts to the rest of that to-do list, and you're not going to shrink the firm size necessarily overnight. There's some areas of the broader global ecosystem where that's not true, those jobs will go away. Customer service rep jobs, support agent jobs, those types of things will be reduced in quantums for sure and in the short term. But if you think about high level thinking type jobs, like being a private equity investment analyst, that job is not going away.

Shiv: Yeah, totally agree. I don't think a lot of those core jobs can actually disappear. I think over time there will definitely be margin expansion. I think we're kind of in this messy middle where people are still trying to figure out what is the right balance of resourcing versus focus and where is the real value creation opportunity. So I understand that. Help me understand, again, I'm kind of exploring this out with you just based on the product that you've built here. What about on either the PE firm side or the company side? Again, coming back to the product that we've been building Inquisio, one of the reasons we built it is that we felt like we had a proprietary data set. And I often talk about this concept of general AI tools versus verticalized AI tools. And verticalized AI tools have this advantage of a proprietary data set and a proprietary algorithm that generally AI tools just can't have access to, right? So how do you see AI playing a role in the PE space in that way? For us, it's been that we have worked with hundreds of PE firms and we have proprietary marketing data on thousands of companies that we've leveraged to build our own proprietary platform. But I would imagine inside PE firms, are millions and millions and terabytes of basically different data points that can be leveraged to build AI tools or make their solutions more AI enabled to capture some of this value. What are some things that you're seeing as patterns in terms of value creation or just strategic avenues that these companies should be exploring or are already exploring?

Raj: Yeah, I think, so sort of going to your first point of horizontal versus vertical solutions, couldn't agree with that more, we're vertical solution obviously, so that's core to our thesis. The reason we think that's core to our thesis is what you mentioned around the proprietary data. That data is a treasure trove, right? Terabytes of past deals of portfolio company data, of IC memos and diligence reports, market studies, expert network transcripts, leveraging all of that data can empower decisions. A lot of these firms don't even realize they're sitting on that data. And especially the younger associates and analysts who weren't around when that deal happened have no idea that even exists. So being able to leverage that data, talk about margin expansion. That margin expansion is not a cost savings margin expansion. That's a do better deals, margin expansion for them. Now we get anything a 5x MOIC instead of a 3x MOIC because we're smarter about the way we approach things. So I think that's absolutely paramount. 

The internal data moat, different firms have different levels of that internal data. think people are recognizing that one, you can't train custom models on your data, but you can leverage it for the way that you're going to act for GoForward. And the future of this, and some firms are already doing this, is that you will have agents that act as part of your investment committee and vote on investment committee. And they're actually going to provide rationale and reasoning behind their votes. People use the word agent in all sorts of different ways. I think it's actually, frankly, become a marketing buzzword at this point. Half of what people are saying are agents or not actually agents. But the agents that we expect to exist within private equity firms are acting with either autonomous or semi-autonomous decision making. They're connected into data and they're executing multi-step processes for the firm. In the example of the IC committee member, right, the agent that is an IC committee member, it is going and doing a whole bunch of research on a deal, looking through a data room, understanding past IC memos, understanding the risks of that deal, doing broader market research, combining all of that to project out a view on whether this deal makes sense for the firm. That decision, interestingly, is not the same for every firm. So if you gave that agent to firm A and to firm B, you'll actually end up with different outcomes because those two firms have different operating teams to support that portfolio company. So that agent needs to understand we have really strong expertise in this domain on our operating team and therefore we believe we could get better returns than know, firm B might. That's a future of where this is going and you can't do that with a horizontal solution. It's never going to have that level of understanding. This is our firm. This is what we've done. This is how we've we've operated in the past.

Shiv: So do you envision a future where a lot of these firms have multiple verticalized AI tools that are part of the deal or investment cycle as they’re vetting companies?

Raj: We're thinking about agents as effectively members of each department, and you might have multiple of those. So if you go back to the sourcing example, there's going to be a sourcing agent. That sourcing agent is going to have an identity and a persona. You might have a name for them that you refer to them like a team member. They'll be on your group chats. They will have email addresses. The bankers will send their decks to that agent. That agent will then know to automatically parse out that deck and do work against the CRM and do work against the internet, figure out who to email and try to schedule meetings with those people, that's gonna be an autonomous or semi-autonomous agent. The biggest headwind, that's actually all possible today. The biggest headwind on that is actually risk and compliance, right? Making sure that we're doing this in a way where it's on tight guardrails, it's not accidentally sending out information that shouldn't be sent out, that is not sending out erroneous emails or putting the firm's reputation at risk. That's actually the biggest thing that we're all trying to solve for right now.

Shiv: Mm hmm. Yeah, I totally see that as well. I think the every area of value creation will have its own verticalized tool set. And that can be from sourcing to marketing to sales to the pricing. I think pricing is a huge area to even M&A, just how are you going to add on or bolt on acquisitions to a platform or cross sell and upsell. We've seen that as a huge opportunity where you're looking at whitespace opportunities. And there's just so much customer marketing that can happen or customer cross-sell upsell that can happen inside platform companies and nobody's looking at that internal data even. So I think a big, starting point is that I think that's kind of being overlooked and I want more PE firms to kind of think about it is are the internal data sets or proprietary data sets that should be analyzed? You don't have to go build an agent, agentic application off of that. But just starting with what are all the proprietary data sets and analysis that we need to do and how can that be leveraged so that we create more enterprise value from that.

Raj: Yep, that's exactly right. And we think about it in two ways. It's what can we do to make us faster at what we do today? And what can we do to make us smarter? And leveraging that data is often then a second bucket of how can we be smarter about how we're interacting with our portfolio companies or guiding them?

Shiv: Yeah, and as we're coming up on time here, I'm just curious, you know, as tons of PE firms listen to this, like, what would your advice be in terms of what, should they go about incorporating some of this into their day to day operations, or how they think about building their firm or looking at their, at their portfolio companies? What are some initial steps that they can take?

Raj: So first off, a lot of firms have already started trying to skin this cat and understanding the opportunities that are out there. So talk to your peers. Your peers have started down this journey. If you don't have an AI committee yet, go launch that AI committee. Generally recommend that committee is actually cross-functional. So bring on folks from deal teams across different sectors, if you have different pods or sector units. Bring on folks from operating or portfolio value creation teams, bring on folks from IR and fundraising and back office, have compliance, have a representative and so on. Once you have that group together, compiling the list of use cases across the firm, building that inventory of possibilities is the first thing to do. And then stack ranking those is the second thing. Almost inevitably, investment team use cases and portfolio value creation use cases are gonna bubble to the top. Then think about sort of the fundamental ways that you're going to go about tackling the list. What do you need to get it done? Can you do it with a horizontal solution? Do you need a vertical solution? Do you have the data? Do you have connectivity to the data available to you that you need to execute on that? Once you do that, build a project plan, assign ownership. A lot of people don't have someone to give this to. I often say this works best when someone's promotion hinges on it. This is their project. If they get it done and get it done well, they have a really great path at the firm because the stakes are very high. The value is really high if you get it done well. But if you don't assign accountability and ownership to it, it's going to linger. People get distracted by deals. They get distracted by what's going on with portcos. Things just don't get done. So make sure you have someone who's owning it and driving it forward.

Shiv: Yeah, that's awesome. And for the folks that are listening, if they want to learn more about you or BlueFlame, what's the best place to learn more about that, to learn more about the product and just connect with you.

Raj: Yeah, I appreciate that. We're online at blueflame.ai. My email address is raj at blueflame.ai. I'm always happy to field questions, ideas, help with anything.

Shiv: Awesome. Yeah. Thanks for sharing that. We'll be sure to include that in the show notes and Raj with that said, appreciate coming on and sharing your expertise. I think, you know, AI is just one of those topics that a lot of private equity firms are thinking about right now. And I appreciate you coming on and sharing your expertise because I think there was a lot of great insights for firms to take. So appreciate you doing this.

Raj: Thanks for having me, great questions. Thanks, Shiv.