73 Strings Launches Extraction Agent: Agentic AI for Private Markets Reporting

73 Strings has launched Extraction Agent, an AI-driven solution designed to automate the extraction and mapping of financial data for alternative asset managers. By replacing manual data entry with a multi-agent workflow, the tool addresses the operational bottleneck of processing inconsistent reports, allowing investment teams to accelerate portfolio reporting while maintaining high levels of auditability.

What was announced

73 Strings introduced Extraction Agent as an integrated component of its 73 Extract platform. The solution is specifically engineered for private equity, venture capital, and private credit firms that manage diverse portfolios where reporting formats vary significantly between companies. It utilizes a multi-agent AI architecture, featuring specialized agents—including concierge, intake, extraction, mapping, and validation agents—coordinated by an intelligent orchestrator.

The system combines deterministic extraction with an agentic AI layer to ensure both consistency and adaptability. Key features include "Soft Memory," which allows the AI to learn from user corrections and apply those preferences to future documents. It supports various file types, including PDFs, spreadsheets, and scanned images, and can extract data from tables, charts, and commentary across multiple languages without the need for rigid templates. Crucially, the tool maintains source-level traceability, linking every extracted figure back to its original location in the source document.

To ensure enterprise-grade security, 73 Strings has implemented a zero-data-retention policy, meaning no client data is used to train the underlying AI models. This "AI-as-maker, human-as-checker" approach is intended to shift investment professionals away from manual data entry toward a review-based oversight role, requiring human sign-off before any data is finalized. The platform supports use cases ranging from extracting EBITDA and debt figures to automating borrower reports for ongoing credit monitoring.

The companies involved

73 Strings is a fintech firm specializing in AI-powered valuation and portfolio intelligence for the alternative assets sector. The company serves a global client base of asset managers who require sophisticated tools to manage complex, non-public financial data. The firm has attracted significant interest from major financial institutions and investment groups, positioning itself as a specialist in digitizing the middle and back-office functions of private market firms.

The broader ecosystem surrounding 73 Strings includes notable names such as 7RIDGE, a specialized growth equity firm, and Broadhaven Capital Partners. Other major players in the fintech and private markets space that intersect with this segment include Hamilton Lane, a leading private markets investment firm, and global banking giants like Goldman Sachs and Golub Capital. Blackstone Innovations Investments has also been linked to the firm’s development. These associations highlight 73 Strings' position at the intersection of traditional private equity expertise and modern machine learning applications, aiming to replace the manual, spreadsheet-heavy processes that have historically defined the industry.

What FF News has reported before

FF News has previously tracked the activities of 73 Strings and its partners as they expand their reach in the private markets. This includes coverage of Chubb Wealth and Hamilton Lane Partner to Unlock Private Markets for Hong Kong Investors, which highlights the growing demand for private market access and the infrastructure required to support it. Additionally, the publication has reported on the technological shifts at major institutions, such as when Goldman Sachs Goes Live on Equilend 1Source Distributed Ledger for Securities Finance. These developments underscore a broader industry trend toward automating complex financial workflows and improving data transparency across the global investment landscape.

What this means

This move by 73 Strings signals a shift from simple OCR to "agentic" AI in the back office. For years, the alternative asset industry has been plagued by the "unstructured data problem," where every portfolio company reports differently. By introducing a system that learns from human corrections, 73 Strings is putting pressure on traditional fund administrators and legacy software providers who still rely on rigid templates. The real test for the sector will be whether these AI agents can truly handle the nuance of complex credit agreements or idiosyncratic reporting without significant human intervention. If successful, it could drastically reduce the reporting lag that currently frustrates Limited Partners.