Subscribe
Share
Share
Embed
The "manual era" of capital allocation is in its final chapter. The firms still relying on manual data extraction and analysis aren’t failing overnight, but they are falling behind one week at a time. While most of the industry continues to "white-knuckle" through 200-page documents and legacy databases, and manual Excel extraction, a new breed of Agentic Allocators is quietly rewriting the rules. They aren’t just using AI to summarize emails; they are leveraging AI-augmented workflows that intelligently automate parts of their investment and operational processes that were previously impossible to automate.
Hosted by Victoria Sienczewski, CEO and Founder of AuumAI, The Agentic Allocator is the "behind-closed-doors" look at how the world's most sophisticated Limited Partners (LPs), allocators and General Partners (GPs) are actually deploying AI, and the hard-won lessons from those building the systems.
This isn't a series about high-level theory or technical gibberish. Each conversation features industry leaders, forward-thinking LPs, GPs and experts who are rewriting the rules of capital allocation through agentic AI. Expect real-world case studies, tactical frameworks you can actually use, and moments that challenge outdated norms. You'll come away with a clearer understanding of the critical questions every allocator must ask - about data privacy, team adoption, integration, and governance - before investing in any AI solution. If you're tired of the "black box" and ready to evolve your investment office for what comes next, you're in the right place.
The manager research team job will inherently change. It'll be back to a people business. As much as social media hit us behind a screen and led us to believe these are real things and this is the real world, I think AI is gonna shift that back to a world of, I actually need to see it firsthand.
Victoria Sienczewski:Welcome to The Agentic Allocator. Today, we're joined by Ryan Kulig from SITFO, the Utah School and Institutional Trust Fund's Office, a $4,700,000,000 permanent fund for Utah's public education programs. For the past sixteen months, Ryan has been conducting a rigorous AI landscaping exercise, evaluating solutions, building agentic workflows, and integrating AI into the operational fabric of SITFO.
Ryan Kulig:I think having sound support and buy in from the management and and the board is super important on the get go. And then once you have that, then empowering the team to explore and find solutions and acknowledge where there could be automation or agentic workflows in place.
Victoria Sienczewski:As a member of various industry networks of leading endowments, foundations, and health systems, he found virtually no one else was actively implementing AI, which only deepened his conviction to lead. Today, we find out how.
Ryan Kulig:Another challenge has been integrations, how we get the piping across our CRM, our benchmarking, our performance vendors, how we link it to our g Suite, our emails. Getting those dialed in and permissions controlled has been a challenge, but I think in the end, will undoubtedly make AI so much more powerful.
Victoria Sienczewski:I hope you enjoy the conversation. Ryan, thanks so much for taking the time to speak with me today.
Ryan Kulig:Yeah. It's a pleasure. Thanks for having me.
Victoria Sienczewski:Ryan, we'd love to start off before we dive into AI, our topic for today, to go back to the inception of SITFO. Can you walk me through how you grew the team and grew the organization?
Ryan Kulig:Yeah. It's, you know, time flies. It's been almost ten and a half years since SITFO started. We started with three people, really. Peter, our CIO, was the first hire, and then myself, and one of the other investment analysts were the next hires.
Ryan Kulig:So it's been quite the adventure. I remember early days, we were sitting down with our board members, establishing the foundational documents, our investment policy statement, statement of investment beliefs, and I remember it was painful sitting around the board table and wordsmithing line by line those documents, but those were foundational to our agency, so it was important to get those out of the way. But we inherited a portfolio, basically of seventythirty Vanguard mutual funds and some core real estate, value add real estate partnerships. So from there, we hired an investment consultant, a custodian bank, you know, got the foundations established, and off and running we went.
Victoria Sienczewski:Incredible. Would maybe love to start from the data foundation, if you don't mind, and hear how you thought about integrating all of the different kind of systems in your tech stacks, you're building that context layer for your Agentic AI applications.
Ryan Kulig:Yeah. We really started kind of naively where we thought, hey, can we use AI to generate investment on those for us? And then, you know, early on, we quickly found that was a commoditized solution. And so we're like, what can we do to make it broader, cross functional capabilities across the organization? We have really three verticals, finance and operations, strategy and risk, and manager research.
Ryan Kulig:And so can we find an AI solution that can service all of those verticals across our our agency? And so that led us to really explore broader products, and one of the key factors we wanted to look at was how does it integrate with our already established relationships and systems and tech stack as it is, rather than another platform to log in so the team has, you know, four or five systems log in to. Why can't we just get credentials for one and have it integrate and overlay across all those systems?
Victoria Sienczewski:Phenomenal. It's so important. That's what I think about every day, creating that kind of digital intelligence foundation or that context layer so you're able to leverage Agentic AI to its fullest extent. Before we dive into some of the specifics of what you've built, you're part of a variety of different industry groups, so leading endowments, foundations, health networks. And you mentioned that almost no one within those communities was actively implementing AI.
Victoria Sienczewski:Curious, what do you think is holding the industry back?
Ryan Kulig:Yeah. I think, you know, institutional investors are inherently risk conscious, meaning we're fiduciaries for a permanent pool of capital, and so we're typically slow to adopt new technology or evolving technology until it's proven in the marketplace. I also think that institutional investors are viewing AI as a productivity tool, rather than a transformational technology. And so I think once they understand, you know, the enhancements across workflows or data sets or operating models, I think they'll truly realize value there. And then lastly, I think AI implementation is hard.
Ryan Kulig:And so getting your datasets organized, data cleaned, and everything ready to be implemented or utilized in the AI infrastructure, I think that's kind of scary for people. And so, at SITFO, we viewed the cost of being earlier outweighed the risk of being earlier in the space, and so we really wanted to hit the ground running and get something implemented across our tech stack early.
Victoria Sienczewski:Amazing. And it you hit on so many important points around industry adoption. So fully agree with you. Right? Allocators and LPs have historically been slower adopters of technology.
Victoria Sienczewski:What's interesting for me as I'm having numerous conversations and working with LPs and allocators is that in many cases, the LPs and allocators that have, let's say, a commercial reason, right, a competitive reason to adopt AI. They're moving faster if they have underlying clients, for example, OCIOs or fund of funds. But you were extremely forward looking at SITFO, and you mentioned, right, the benefit outweighing the cost or the risk of implementing this technology as it's still continuing to develop. Maybe you can dive into that a little bit more for us, and to walk us through how you're thinking about AI implementation.
Ryan Kulig:Yeah. As I said earlier, we really are looking at it as a cross functional capability, and we chose the route to find a solution that could benefit every group in our agency, every vertical in our agency. And so part of what we're implementing is landscape maps for manager research, as well as transaction reconciliation for the finance and operations department, as well as risk reporting or performance reporting from the strategy and risk team. And so we truly believe what we're building is gonna benefit the agency as a whole rather than one specific vertical.
Victoria Sienczewski:Fascinating use cases. I love that you're looking at it as an end to end, almost operational transformative opportunity for SITFO and focusing on different use cases across different verticals within the organization. Can you dive into some of these use cases, what you're working on right now?
Ryan Kulig:Yeah. I I think the low hanging fruit is obviously document intensive processes. So whether that be reviewing limited partnership agreements, or populating subscription documents, or redlining NDAs, those are all low hanging fruit for us. And not to eliminate the human aspect of things, but to really enhance it. And so I got feedback from our attorney recently that, oh, we're not gonna replace the attorney, and I agree, but having the basics redlined in an NDA makes that attorney or empowers the attorney to actually dive deeper into, you know, the document, into a more thorough review as they're no longer having to focus on the the baseline red lines of a document.
Victoria Sienczewski:Fantastic. So you mentioned LPA analysis and some of the low hanging fruit. You also mentioned, obviously, the vision, right, of what you're building is to focus on the value add for SITFO over time. What do you think those use cases will look like? Maybe you're already working on some of them.
Victoria Sienczewski:How do you think that they'll look like for you?
Ryan Kulig:Yeah. I think it's really exciting. I think there's a lot of really exciting use cases for our agency. Something we're really proud on that we built recently is to enhance our pipeline capabilities. So top of funnel, filtering down to bottom of funnel across the different asset classes.
Ryan Kulig:And so we built a tool that links to our CRM, or is integrated with our CRM through an API. It will pull documents daily from the CRM system, whether that's call notes or or new documents uploaded, and then file them in either an active manager hub or a prospective manager hub. And from there, we can landscape the universe of every touch point we have in our network. And so, using buy out an example, we can have all our buyout managers in a landscape map, and then screen them based on desirable metrics for the asset class, and how it aligns with our strategic thinking or our thesis. And then, as we choose to move forward in the process, get documentation output, as I mentioned earlier, commoditized product.
Ryan Kulig:And so we think that's truly a powerful tool to enhance the bandwidth of our team so they can cover more with less.
Victoria Sienczewski:Absolutely. I'd love that use case. How do you widen the top of funnel as wide as possible and then allow you to filter down as much as possible to get to managers that really are a fit for your organization? Such a wonderful use case, and I think will add a lot of, hopefully, positive performance outcomes for different organizations as you're able to see more and also benchmark better. That benchmarking or performance analysis at the tips of your fingers is extremely important.
Victoria Sienczewski:Obviously, there's a huge cultural transformation as part of this as well. Maybe you can walk us through, Ryan, how have you thought about the cultural transformation, and how is the team involved in building this solution?
Ryan Kulig:You know, the inherently, I think the job's gonna change of investment manager research. We're no longer hiding behind a screen, evaluating track records and the quantitative stuff. That's all easily done by AI. We want to focus on relationship building, getting to know the managers, getting to know how they act in stressful environments. How is their culture at their office?
Ryan Kulig:How is their team camaraderie, etcetera. And so, I think getting our analysts from, you know, drafting memos to now spending time doing reference checks, getting to know who the manager is as a person, will inherently make our process even more robust.
Victoria Sienczewski:Makes perfect sense. Would love to kind of zoom out and also talk to you about how you view the landscape evolving with AI. Curious, how do you think your office at SITFO will look like in three to five years with AI?
Ryan Kulig:Yeah. I think it will enhance the team. It'll enhance our capabilities. It'll enhance the opportunities we see. Hopefully, enhance returns along the way.
Ryan Kulig:But I think for our team, the two most recent hires we've made have been more technical oriented rather than specific industry match, or manage a research background, or anything like that, because I think we're gonna spend a lot of time coding, developing new prompts, enhancing AI capabilities to really lay that AI framework below the agency to enhance everybody's skill set and job responsibilities.
Victoria Sienczewski:It's fascinating to hear how you're thinking about the shifting role of investment professionals in an AI world. Curious, what do you think that job will look like in a couple of years?
Ryan Kulig:I think, as I was saying earlier, the manager research team job will inherently change. It'll be back to a people business. As much as social media hit us behind a screen and led us to believe these are real things, and this is the real world, I think AI is gonna shift that back to a world of, I actually need to see it firsthand. I need to experience it firsthand to see if it's true, because AI can make anything fake. And so I think that's gonna change the way investment manager research folks will interact with people.
Ryan Kulig:Again, less so memo writing and track record analysis and more face to face interaction and getting to know a person firsthand.
Victoria Sienczewski:Yeah. And and also being keenly aware of, let's say, AI washing, right, on the GP side as well. LPs and allocators will start to use AI more, but also the other side of the ecosystem well and being able to see which managers use it in a really productive way to enhance their investment philosophy and strategy and approach versus others who'll which maybe make it sound better better that it is. Very, very interesting. And we'd love to shift focus to also ask for your advice for other LPs and allocators that are earlier on in the journey.
Victoria Sienczewski:Would love to maybe first of all start with when you're thinking about looking at an AI solution to adopt, what are some of the key components that you would suggest your peers look at?
Ryan Kulig:I think first and foremost, this is specifically true of institutional allocators that what are the resources available to you, whether that's team or bandwidth or skill set or monetary components? What what do you have to work with? And second, what are you looking to achieve? And I think that kind of dictates a path you can go down to. More of a customized solution if you have a lot of resources and highly technical skill set on the team, or more off the shelf offering that the team can easily implement and run with.
Ryan Kulig:And so, I think those are really two components that get you started, and then from there, landscape. We talked to about 15 different vendors in our process of running our RFI, slowly came to two finalists after landscaping, and I think spending the time with each of the vendors, understanding the business, the product offering, the asset class coverage, security protocols, and do those align with what you're looking to establish or create with Agentic workflows? I think all those are components that I would look to evaluate.
Victoria Sienczewski:Fantastic. And would love to also ask for your advice for your peers, LP and Allocators, on the governance side. Implementing and being a leading force within the this transformation takes quite a lot of governance work as well. Curious if you have any recommendations on the governance front.
Ryan Kulig:Yeah. I think we're fortunate where we have a very supportive board, and we have Peter's a a great leader in forward thinking about AI. We actually had established a quote unquote Skynet group about three years ago to really explore how to implement AI across the organization. So this has been a lot of time in the works, and so I think having sound support and buy in from the management and the board is super important on the get go. And then once you have that, then empowering the team to explore and find solutions and acknowledge where there could be automation or agentic workflows in place, I think, is super important.
Victoria Sienczewski:And, Ryan, final question to wrap up. Obviously, the path of AI adoption isn't without its challenges. Would love to understand what challenges have you worked through?
Ryan Kulig:I think with any new system, data integrity is a huge challenge. And so we spent a lot of time cleaning our CRM, making sure that the tags are correct, and we've instilled audit process and anything that goes into the CRM, and that way, the workflows can come across, or the documents can come across accurately and acknowledged by the AI accurately. And so I think that was a huge challenge for us. Another challenge has been integrations, and how we get the piping across our CRM, our benchmarking, our performance vendors, how we link it to our g Suite, our emails, shared emails, G drive, our network folders where we store all of our data. And it's important to consider permissions there.
Ryan Kulig:We don't want AI to taking my personal email and sending out emails on my behalf, and because that could deal with sensitive information such as, you know, wire instructions, updating those with the managers, and so on. And so getting those dialed in and permissions controlled has been a challenge, but I think in the end, being able to foster underlying data that emails and network folders and all the other integrations have will undoubtedly make the AI so much more powerful.
Victoria Sienczewski:It's absolutely critical, but not necessarily the most sexy or funnest part of the transformation journey. So kudos to you to having implemented such a fantastic use case internally. Ryan, thank you so much for taking the time to join me today.
Ryan Kulig:Thank you, Victoria.
Victoria Sienczewski:That's a wrap for this episode of The Agentic Allocator. If today's conversation gave you a clear vision of where the industry is headed or helped you pinpoint exactly where your own process is stuck, go ahead and follow or subscribe wherever you get your podcasts. Your firm's judgment took years to build. Stop letting it reset every time a team member walks out the door. Visit auumai.com to discover the structured digital intelligence foundation your team needs to leverage Agentic AI to its fullest potential.
Victoria Sienczewski:Until next time.