Use Case Intelligence RSS Search use cases Search
OpenAI Financial Services North America

Benefit Street Partners

Document Processing · Risk Management

Benefit Street Partners needed to answer investment-committee follow-up questions faster, compare lender marks across shared loans, and review 300-page credit agreements for clauses that could shift value away from lenders without waiting on slow manual analysis or outside counsel to read every page.

Read the original story Explore all use cases
10–12 hours total.
The first version took a day.
about 20 minutes to process into a first-pass analysis
an hour or more of human review

Reported impact

Business outcome

Benefit Street Partners can now turn a round of investment-committee comments in a day, often answer follow-up questions during the meeting itself, build the first version of its BDC mark tracker in about a day, and generate a first-pass credit agreement analysis in about 20 minutes before human review, adding more review capacity and confidence within the same underwriting window.

  • Productivity Improvement
  • Risk Reduction
  • Scalability

Solution

Benefit Street Partners built two AI-assisted underwriting tools around Codex and GPT-6 Astra. For loan valuation work, portfolio manager Anant Kumar created a BDC mark tracker that pulls public SEC filings from business development companies and compares how different firms value the same loan, giving BSP a dashboard for benchmarking its own marks. Kumar used Codex to draft and review the application code, then iterate on features and interface changes in hours. For document analysis, BSP assembled memos from leading law firms on clauses and lender risks, then paired that reference material with uploaded 300-page credit agreements inside a credit agreement analyzer. After BSP refactored the tool with GPT-6 Astra and put it in a company-hosted portal, the system could ingest long agreements, synthesize a first-pass risk review, and hand the result to analysts for more than an hour of human validation before any investment decision.

Data flow

Public SEC filings feed the BDC mark tracker so BSP can compare its loan marks with other holders' valuations. Separately, analysts upload long credit agreements and pair them with law-firm memos on risky clauses inside the credit agreement analyzer, where GPT-6 Astra produces a first-pass review in the company-hosted portal. Analysts then review the output before using it in underwriting rounds and investment-committee follow-ups.

Solution architecture

4 components · 2 layers
  1. Compute
    • Codex Helps Kumar draft, review, and refine the code behind BSP's internal underwriting tools.
    • GPT-6 Astra Powers the refactored analyzer that ingests long agreements and synthesizes the initial review inside BSP's portal.
  2. Serving
    • BDC mark tracker Compares how business development companies value the same loan so BSP can benchmark its own marks against other holders' valuations.
    • Credit agreement analyzer Processes uploaded credit agreements and surfaces clause-level issues and lender risks in a first-pass underwriting review.

Architecture clues

  • BSP assembled memos from leading law firms on clauses and risks to watch for as reference material for the analyzer.
  • BSP is exploring continuous underwriting so it can monitor loans as news happens across the portfolio.
  • BSP refactored the analyzer with GPT‑6 Astra and deployed it in a company-hosted portal.
  • Human review remains part of the workflow before any investment decision is made.
  • The BDC mark tracker uses public SEC filings from business development companies to compare how different firms value the same loan.
  • The credit agreement analyzer works on agreements that are often 300 pages or longer.

Evidence from the source

"An uploaded credit agreement takes about 20 minutes to process into a first-pass analysis, followed by an hour or more of human review."
"BSP has refactored the tool with GPT‑6 Astra, and it is now live in the company-hosted portal."
"His tracker uses public SEC filings from business development companies (BDCs) to compare how firms value the same loan."
"Kumar built the tracker and used Codex for code review."
"We feel more confident in our decision making because we have more information at our disposal."
"answering an investment committee’s follow-up questions used to take days or weeks."
Transparency Evidence & methodology 100/100 reference value · 95% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    2 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence95%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability25/25
    12 of 12 key fields link to a named source.
  • Direct evidence20/20
    6 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency10/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. OpenAI

    Benefit Street Partners answers questions faster with Codex | OpenAI

    First-party
    https://openai.com/index/benefit-street-partners/
    Source type
    Vendor-published customer story
    Published
    Oct 16, 2026
    Supports
    • Customer
    • Industry
    • Region
    • Company size
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • Business outcome
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Oct 16, 2026 · OpenAI · First-party
    Open date source

Related

Similar use cases

See all like this
OpenAI Nov 10, 2025 BNP Paribas Financial Services global Relationship Managers needed hours to complete ESG Assessment questionnaires of up to 80 questions from client ESG and CSR documents, creating operational inefficiency and data quality issues in a process embedded in… 600 users since the Go live Document ProcessingRisk Management AI Assistant · Azure API · ChatGPT 4o mini OpenAI Dec 4, 2024 Endex Financial Services United States Investment professionals needed AI that could reason through complex financial documents and data with the precision, traceability, and multistep thinking required for real analysis rather than generic search results. Financial experts preferred responses generated by OpenAI’s o1 model 70% of the time Document ProcessingRisk Management AI Analyst · GPT-4o · o1-mini OpenAI Oct 15, 2026 Sullivan & Cromwell Commercial and Professional Services North America Sullivan & Cromwell needed to review draft agreements against relevant precedents, client priorities, and market practice in one workflow so lawyers could do deeper legal analysis without losing transparency into… Document ProcessingRisk Management Agreement Analyzer · ChatGPT Enterprise