Underwrite · PCT-040 · Sourced card
SRT reference pool check
A credit fund is offered the junior tranche of a bank's synthetic risk transfer on a €2bn corporate and project-finance reference pool with a replenishment period.
Today vs automated
- Read eligibility rulesAssociate · manual
- Profile the tapeAssociate · manual
- Guess the worst caseVP / counsel · manual
- Email / meeting hand-offAssociate → VP / counsel
- A bank sends the reference pool and offering termsTrigger: starts on its own
- Extract replenishment rulesAutomated · Document parser
- Build the worst poolAutomated · LLM
- Model tranche lossesAI draft, human review · Model template
- Price and participationApprove / override · VP / counsel
AutomatedAI draft, human reviewHuman onlyTools: Data room, Document parser, LLM, Model template
Estimated valueestimate
30–360 analyst hours a year
$4k–$90k of time, vs an assumed $10k–$40k tool cost
Net: −$36k to $80k a year net of tool cost; at the low end of the assumptions it does not cover the tool cost.
Assumptions
- 10–20 analyst hours per pool today; 2–5 hours with an agent, including review
- 6–20 pools a year for one team
- $150–250 per hour loaded cost
- Tool or build cost $10,000–40,000 a year (in-house)
Estimate, based on the assumptions shown; your numbers will differ. How we estimate
What else it can do for the team
- Profile pool by sector, rating and size
- Track replenishment against limits
- Draft the investment committee summary
- Compare pools across bank issuers
- Monitor monthly pool reports
How you get thereFour steps from prompting to an agentic workflow; most teams start at step 2.
Needs: an approved AI tool, a confidentiality policy.
Risk: unchecked answers, data pasted outside policy.
Needs: field definitions, a template, document parsing.
Risk: breaks silently when formats change.
Typical starting point
Needs: inbox/doc-store access, system write-back, logging.
Risk: acting on a misread; needs review gates.
The automated lane above
Needs: shared data model, permissions, audit trail, exception queue.
Risk: errors compound across steps; needs monitoring.
Solution optionsBuy, build or hybrid, with the trade-offs of each.
No vendor verified for this workflow in our directory yet.
Fastest start; you adapt to the product's data model and pricing.
An LLM API, a document parser, a vector or document store, your data warehouse, and a review queue.
Fits your definitions exactly; you own maintenance and controls.
Vendor extraction feeding your own checks, templates and systems.
Common middle path; two things to integrate and govern.
Listing is not endorsement.
What changes for the team10–20 h → 2–5 h per pool (estimate); coverage goes from the sections someone had time to read to every clause and schedule checked against the term sheet.
How it works
Full title: SRT reference pool analyzer: what's actually in the pool you're insuring, and what can the bank add later?
“Today's pool is BBB-average. Under the replenishment criteria, the bank can make it BB-minus. Here's the clause.”
Scenario
SAMPLE: a credit fund is offered the junior tranche of a bank's synthetic risk transfer on a €2bn corporate and project-finance reference pool with a replenishment period.
What the human still decides
Price and participation. The agent maps the documents and runs scenarios; the investment committee decides.
Why it matters
Banks are widening SRT collateral: Deutsche Bank is working on SRTs referencing ~$2bn of subscription lines and ~€2bn of project finance including data centers; ING agreed a €5bn project-finance risk-sharing deal with PGGM. The FSB notes private credit funds are the main SRT buyers in Europe.
What the record shows
Checked 2026-10-15.
Reference pools are moving into private-credit-adjacent assets. ING and PGGM's Linden-1 references €5bn of project-finance loans (reference portfolio, not risk transferred; tranche and pricing undisclosed), and Deutsche Bank is working on an SRT referencing about $2bn of subscription lines, with its CFO citing 75–80bp of CET1 relief from SRTs at end-Q2. A protection seller needs the pool's sector, sponsor and vintage mix, which these announcements do not give.
Try it: prompt and data
From this SAMPLE SRT term sheet, extract eligibility criteria, replenishment conditions, concentration limits and credit event definitions with cites. Then construct the weakest portfolio permitted by those criteria from the SAMPLE obligor universe and compare its rating and sector mix with the current pool.
Data
Method shown on SAMPLE data. Real portfolios, trustee reports and data-room files are confidential: run on approved internal tools only, never paste into a public AI tool.
Who uses it, vendors and build
- Segments
- Bank partnerships / SRT
- Who uses it
- SRT investor (credit fund), Structured credit analyst, Bank capital-relief team
- Build
- Buildable in-house
- In-house
- Clause extraction plus a constrained portfolio optimizer. Estimate: 3–4 weeks.
- Last verified
- 2026-10-15
Vendors that claim this capability
- No vendor claim verified for this workflow yet.
Tags reflect each vendor's own public claims at the linked page (checked 2026-10-15). Not tested, ranked or endorsed.
Sources and related
- BCR: ING and PGGM €5bn project-finance risk-sharing deal
- American Banker: Deutsche Bank plans risk trade tied to $2bn of fund loans
- FSB: Report on vulnerabilities in private credit
Part of the deal lifecycle: stage 2 of 7, Underwriting.
Next step
Get new walk-throughs
Not investment advice. Names marked SAMPLE are fictional. Found an error? See the corrections log.