About a-team Marketing Services
The knowledge platform for the financial technology industry

A-Team Insight Blogs

S&P Builds Private Markets-Trained AI Document Search Tool for iLevel Platform

Subscribe to our newsletter

S&P Global Market Intelligence has expanded its private markets data and technology platform iLevel with the addition of AI Document Search, a module that is built on large language models (LLMs) trained specifically to aid participants in the fast-growing alternative assets sector.

The new tool enables general partners (GPs), who manage funds on behalf of limited partner (LPs) investors, to interrogate the data they hold on their assets using natural language prompts. The innovation is aimed at helping asset managers and limited partners (LPs) make better-informed decisions without having to spend hours poring over financial documents issued by their portfolio funds and companies.

Chris Sparenberg, S&P Global Head of iLEVEL, said Document Search reduces the manual work clients go through to obtain insights.

“We provide document and data collection services to iLEVEL clients and we collect a tremendous amount of information on their behalf,” Sparenberg told Data Management Insight. “Due to the volume of unstructured data in their environments, iLEVEL users still need to access underlying documents to analyse sentiment, trends and other non-standard information.

“Document Search is a way to remove some of those steps from the process and reduce the need to go and manually review documents. Clients want more data. They have access to it, but paradoxically, it’s locked in PDFs. And so this is the way to bring all of that universe of important information to their fingertips.”

Data and technology vendors are clamouring to bring the sort of transparency to private markets as investors are used to in public markets as they seek to fill an information gap to a sector that Preqin estimates will account for more almost US$25 trillion by 2028.

Rising Demand

Startups like Juniper Square and Alkymi are jostling with established data companies including CUSIP and NeoXam to serve the new influx of investors eager to capitalise on the higher returns in private equity, private credit, venture funds and property as volatility plagues mainstream capital markets.

iLEVEL was introduced in 2008 as one of the first products to bring transparency to private market investors, beginning life as a template-based tool that LPs and GPs could send to portfolio companies in order to obtain critical investment-decisioning data. It has since been enhanced into a multi-faceted platform that helps both sets of market participants better manage and monitor their investments.

S&P Global already offers a search feature in iLevel that’s largely focussed on rules-based inquiries of entity and document names. Sparenberg stresses that iLevel Document Search’s differentiator is its LLM, which he said has been created especially for the one-niche asset types.

Proprietary LLMs

Clients can use Document Search by entering a query within the platform and the LLMs – which are based on Anthropic’s Claude model suite – queries all of the documents in the company’s own ring-fenced iLevel data store to generate a response.

“The biggest difference is a lot of LLMS will be generalist in terms of their capabilities,” he said from his office in Atlanta, Georgia. “Ours, thanks to our internal expertise and that of the clients who use our platform every day can be smarter and are actually built for the purpose of asset management and portfolio monitoring. It’s the quality of the queries that informs the quality of results.”

As well as benefiting from natural language processes capabilities, iLevel Document Search users can take advantage of features that enable the granular annotations of data to make traceability easier by linking every data point to original sources. Sparenberg stresses that tool is permissions based so that access to results is strictly protected to maintain confidentiality of sensitive materials.

With the data gathered via the new search facility, GPs can create their own portfolio models, populating them with all the information applicable to their funds’ underlying assets.

“There are things that don’t come out [of the documents and PDFs] during standard data extraction, and this gives clients a way to, on the fly, query that additional information, ask questions, and do things that enhance their workflows,” Sparenberg said.

LPs Welcome

The service isn’t restricted to GPs. LPs can also use Document Search to query their databases on the funds in which they are invested, gathering insights into performance and poring over GP letters to get a sense of how asset managers’ thinking is evolving.

Sparenberg said that the latest addition to the iLevel platform is part of a broader programme of development for the offering, and that S&P Global would continue responding to the needs of its clients.

“We spent a lot of time actually talking to clients to validate this concept before we started building and what we found was that document and data volume – even though we solved for much of it and the parts of it that can be highly operationalised is what specialise in – it’s still a topic for our them,” he said. “They want to know how they can better marshal all the information they have access to? How can they more dynamically manage the investments that they’ve made? How can they interact with other stakeholders with certainty?”

Subscribe to our newsletter

Related content

WEBINAR

Recorded Webinar: The ROI of Data Trust: Quantifying the Business Value of Data Observability

Data is the fuel that keeps modern financial institutions’ motors running but if that data can’t be trusted then the decisions made based upon it, or the uses to which its put, will be compromised. That’s especially important for data that’s fed into artificial intelligence models. If the data isn’t clean, accurate and complete, then...

BLOG

Twelve Leading Data Lineage Solutions for Capital Markets

The ability to trace the journey of data from its origin to its final report is no longer a luxury but a regulatory and operational necessity. As firms grapple with the intensifying requirements of regulations such as BCBS 239, GDPR and the shifting landscape of MiFID II, the “black box” approach to data management has...

EVENT

Digital Assets & Tokenisation Forum, New York

A-Team Group’s Digital Assets & Tokenisation Summit spotlights how global financial leaders are rapidly embracing programmable tokenised assets and DLT networks to achieve real-time, 24/7 peer-to-peer transactions.

GUIDE

AI in Capital Markets Handbook 2026

AI adoption in capital markets has moved into a more disciplined phase. The priority is now controlled deployment: where AI can be used safely, where it can deliver measurable value, and how outputs can be governed, monitored and evidenced. The 2026 edition of the AI in Capital Markets Handbook examines how AI is being applied...