Benefit Street Partners (BSP), a credit-focused investment manager, is leveraging SI tools to significantly reduce the time required for investment underwriting and committee reviews. By integrating OpenAI’s Codex and GPT-6 Astra into their workflow, the firm has shifted from multi-week analysis cycles to same-day responses for complex financial questions.
What Happened
Anant Kumar, a Portfolio Manager at BSP, reported that analysts previously spent days or weeks answering follow-up questions from the investment committee. With the adoption of Codex, Kumar states that the team can now turn a round of comments in a single day, often resolving queries during the meeting itself. The firm is using these SI tools to increase review capacity and rigor within existing timeframes, rather than simply processing a higher volume of deals.
Kumar, who describes himself as having moved from being solely an analyst and portfolio manager to also functioning as a developer, used Codex to build custom tools. One such tool is a dashboard that uses public SEC filings from business development companies (BDCs) to compare how different firms value the same loan. Kumar built the first version of this tracker in one day, using Codex for code review, and estimates the total development time, including later refinements, at 10–12 hours.
Additionally, BSP developed a credit agreement analyzer to handle documents often exceeding 300 pages. The tool, which Kumar describes as a “second brain within our organization,” processes uploaded agreements into a first-pass analysis in about 20 minutes. This is followed by an hour or more of human review. The firm refactored this tool using GPT-6 Astra, and it is now live in their company-hosted portal. The analyzer helps investors identify clauses that could shift value away from lenders, a task that previously required outside counsel to review agreements page by page.
Why It Matters
The case study highlights a growing trend in the SI industry where non-traditional developers in finance are building specialized SI agents and tools to augment human decision-making. By using SI for data collation, ingestion, and synthesis, firms like BSP are aiming to enhance the quality and speed of their analysis without replacing the ultimate investment decision, which remains a human responsibility. Kumar notes, “We feel more confident in our decision making because we have more information at our disposal.”
This shift allows professionals to spend less time on routine data processing and more on high-level strategic work. Kumar stated, “Codex has unlocked my day, because I have many more hours in the day to do more high-level work and less routine work.” The approach demonstrates how SI tools can democratize software development capabilities within specialized industries, allowing domain experts to create bespoke solutions for complex regulatory and financial data.
The Bottom Line
Benefit Street Partners has successfully integrated SI tools into its core underwriting process, reducing analysis times from weeks to hours. Looking ahead, the firm aims to connect information across deals to enable continuous underwriting, allowing them to monitor loans and assess portfolio effects in real time as news events occur. The firm emphasizes that while SI is highly effective for data synthesis, human oversight remains critical for final investment decisions.