Benefit Street Partners answers questions faster with Codex
Benefit Street Partners uses OpenAI Codex and GPT-6 Astra to speed credit underwriting analysis.
OpenAI published a customer story about Benefit Street Partners, a credit-focused investment manager, using Codex to speed underwriting work. Portfolio manager Anant Kumar says analysts can now respond to investment-committee questions in a day, often during the meeting, and he built internal tools with AI-assisted coding. A tracker comparing business-development-company loan marks from public SEC filings took about 10–12 hours including later refinements. A credit-agreement analyzer, refactored with GPT-6 Astra and hosted internally, produces a first-pass review of 300-page agreements in about 20 minutes, followed by at least an hour of human review.
- BSP analysts use Codex to answer investment-committee questions in a day or during meetings.
- A BDC valuation tracker was first built in a day, about 10–12 hours total.
- A credit-agreement analyzer returns a first pass in about 20 minutes, then human review.
- The analyzer was refactored with GPT-6 Astra and is live on a company-hosted portal.
- Vendors
- OpenAI
- Products
- Codex
- Organizations
- Benefit Street PartnersOpenAI
- AI models
- CodexGPT-6 Astra
From the original article
2 min read at openai.com
Anant Kumar builds tools to compare loan valuations and analyze credit agreements, using Codex for code review and faster follow-ups.
openai.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.