Private credit deal screening that reads every memorandum against your mandate, and leaves the decision to your underwriters
First-pass screen in minutes, not half a day, every deal read.
See one case, screen by screen ↓Software that takes each inbound teaser and confidential information memorandum and gives the deal team a first-pass view of whether it fits the fund: the financials, leverage, add-backs, mandate and portfolio tests, sponsor record and red flags. Private Credit Deal Screening has six agents do that reading the day a memo arrives, and writes a one-page screen with every figure cited to its page.
How to choose AI solutions for lending →
Updated
A direct lending desk gets more teasers and confidential information memoranda than it can read. A CIM runs 80 to 150 pages, with a QoE summary, a model and a process letter behind it, and an IOI due about a week later. So the team reads the first pages of most memos and the whole of very few.
What gets missed is rarely on page one. The sponsor’s adjusted EBITDA is taken at face value and leverage looks a turn lower than it is. The top customer’s contract renewal sits on page 87. A deal that breaks a portfolio limit gets two weeks of diligence before anyone checks. And a pass is never written down, so nobody learns whether it was the right call.
Estimated split for one full CIM with its QoE summary, by hand and with the solution, in hours. With the solution, the remaining time is the underwriter’s own review.
Six specialist agents mark up every memorandum against the fund mandate and its past passes; the underwriter decides.
The Project Lantern data room opens at 07:58 on Monday 12 October: a 118-page CIM, a 34-page QoE summary and the model, with the IOI due Friday. Here is what happens next, screen by screen, in the working solution.
Daniel does not open a dashboard. He opens the Lantern CIM, read on arrival between 07:58 and 08:04 by six agents. Every sentence a mandate test depends on is highlighted green, red or amber, with a numbered margin note naming the mandate clause and the limit. Above it, a fit of 72, the verdict and a strip of all twelve Fund IV tests; beside it, this week’s other memos, each with its own strip.
“Pursue at up to 5.50×, not as asked — 10 of 12 pass”
“Total leverage of 5.3x Adjusted EBITDA of $54.2 million” is highlighted red. The margin note says the fund’s EBITDA is $46.9M, which puts the $285M ask at 6.08×. With past pass reasons on, the note also links to the three industrial-services deals the fund passed above the same limit: Holcroft Testing, Corliss Compressor Service and Tamsin Pump & Seal, with who decided and where each one is now.
“Mandate max 5.75× — fund passed 3 similar deals.” Two are now in trouble.
From the note, Daniel tries half the Delmont Valve synergies, which is what the add-back policy allows once they are actioned. Our EBITDA rises to $48.55M and leverage falls to 5.87×. Still red. At the 5.75× limit the fund could lend $279M, not $285M.
Thirty-one memos came in this week and all 31 were read the day they arrived, 1,914 pages in all. Eleven are in the pipeline, ranked by fit, each showing EBITDA on the sponsor’s number and on ours. The other 20 were passed on the teaser. Kiln, a burger chain, stops at the sector test because restaurants are excluded, and Basalt shows 8.05× on our EBITDA with add-backs at 41%.
The screen shows each agent’s step in order. The intake reader confirms the confidentiality agreement was signed on 5 October. The financials extractor pulls 64 figures, each with its page. The earnings quality analyst sorts the five add-backs and the tester runs all 12 tests. Four red flags follow, each linked to the CIM or QoE page it came from.
“Finished in 6 min 12 s · 6 agents · 192 pages”
The EBITDA bridge takes each sponsor add-back in turn. ERP implementation, founder compensation and the Toledo branch consolidation are accepted, $5.1M in all, each supported by a QoE page. The $3.3M of Delmont synergies (identified, not actioned) and $4.0M of run-rate pricing (not in the twelve months) are rejected. The leverage meter puts the sponsor at 5.26×, the proposal at 5.50× and ours at 6.08×.
Never accepts a synergy or price increase not yet earned. A person may change a treatment, with a reason that is kept.
Ten pass and two fail: total leverage and add-backs (22.9% against a 20% cap). Top customer Ardent Gulf Refining is a watch at 18.4% against 20%. Each test names its source, from the CIM page to the process letter or fund records. The portfolio view shows industrial services rising from 11.6% to 14.4% of invested capital, under the 15% cap, and qualifying assets at 88.9%, above the 70% minimum.
The researcher reads the fund’s own records: six deals with Harrowgate Lane since 2019, four active and two repaid, no defaults, and one amendment, a 2024 covenant reset at Bellmoor Dental Partners where the sponsor put in $18M of equity. Average leverage at close was 5.13×. Comparable industrial-services loans closed at 5.00× to 5.50× and are performing.
The screen writer sets out three options. Pursue at up to 5.50× our EBITDA, a $258M unitranche at S+550 with OID 98.0 and a $90M hold, inside every limit. Pursue as asked, which needs an exception from the investment committee. Or pass, with a note to the banker. Daniel keeps the first. Six first diligence questions and a note to Pellham Stroud are drafted, and nothing is sent until the decision is made.
The screen writer never sends anything and never decides pass or pursue.
Daniel signs his recommendation, recorded as “I have read this screen and recommend it”. Elena Marsh approves spending diligence money on Lantern. Only then does the note with six questions go to Pellham Stroud, and the decision and its reason are kept in Decisions.
“Elena Marsh approved pursue · questions sent to Pellham Stroud”
The same memorandum, seen by the five people who carry it from the inbox to the investment committee — what their week looked like, and what it looks like now.
Six specialist agents mark up every memorandum against the fund mandate and its past passes; the underwriter decides.
Reads teasers, memoranda, banker emails and process letters; names the company, sponsor, banker, dates and the ask; spots the same deal from two bankers.
Pulls revenue, EBITDA, add-backs, capex, working capital and customer concentration from the memorandum, the QoE and the model, each with its page.
Sorts each add-back into accept, haircut or reject under the fund’s add-back policy, and computes our adjusted EBITDA, leverage and interest cover at today’s SOFR.
Runs the twelve Fund IV tests — credit limits and portfolio limits after the proposed hold — and shows the value, the limit and the source for each.
Reads the sponsor’s record with the fund and the fund’s comparable deals, closed and passed.
Writes the one-page screen, red flags, first diligence questions, proposed terms and the note to the banker.
Passes or recommends; Head of Underwriting approves pursue. The agents propose; a named person decides.
Ask about any deal, figure, add-back or sponsor — or ask for a draft. Every answer points to its page, and it opens the screen it is talking about.
Why is Lantern at 6.08× and not 5.26×?
The sponsor’s 5.26× divides $285M by $54.2M, which counts $12.4M of add-backs. Fund IV accepts the three that are one-time and incurred — ERP, founder pay and the Toledo branch, $5.1M. It does not count the $3.3M of Delmont synergies, none of which have been actioned, or the $4.0M of run-rate pricing. Our EBITDA is $46.9M, so leverage is 6.08×, above the 5.75× limit.
Which deals we passed look like Lantern?
Three industrial-services deals the fund passed above the leverage limit: Holcroft Testing at 6.40× in June 2025, now paying in kind; Corliss Compressor at 6.10× in August 2025, which has since needed a covenant waiver; and Tamsin Pump & Seal at 6.20× in February 2026, on watch.
How has Harrowgate Lane done with us before?
Six deals since 2019 and no defaults. In the one amendment, at Bellmoor Dental in 2024, the sponsor put in $18M of equity. Average leverage at close was 5.13× — Lantern as asked would be their highest with us.
Draft the first questions for Pellham Stroud
Drafted six, one for each red flag and one on structure: the Delmont synergies actioned and when; acceptance of the September price increase; the Ardent Gulf renewal terms; days sales outstanding from 58 to 71; why the QoE left two adjustments out; and a unitranche of up to $258M at 5.50×. Nothing goes to the banker until Elena approves.
13 screens from the working solution, on its sample data. Pick one to see it large.
Every limit the deals are screened against is on the Mandate page. Changes are versioned and on the audit trail; version 7 was approved by Marcus Bell, the CIO.
| Setting | Default | Choose from |
|---|---|---|
| Maximum total leverage on our adjusted EBITDA | 5.75× | Steps of 0.25× |
| Add-backs | 20% of adjusted EBITDA | 15 · 20 · 25% |
| Interest cover | At least 1.75× | 1.50 · 1.75 · 2.00× |
| Top customer | At most 20% of revenue | 15 · 20 · 25% |
| Minimum spread | S+500; S+525 above 5.0× | S+475; S+500 above 5.0× · S+500; S+525 above 5.0× · S+525 throughout |
| Single borrower hold | 3.5% — $112M | 3.0% · 3.5% · 4.0% of commitments |
| Acquisition synergies | Only once actioned, at 50% | Never · once actioned, at 50% · once actioned, in full |
| Who approves pursue into diligence | Elena Marsh, Head of Underwriting | Head of Underwriting · CIO |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample fund and deal data · “estimated” = our estimate · “typical” = published figures (Sutton Place Strategies via V7 Labs — 80–100 CIMs reviewed per closed deal; V7 Labs — 20–30% of a CIM covered under time pressure; WorkWise — a first read takes a day or two). People, funds, sponsors, banks and companies named on this page are fictional — characters and sample data in the working solution.
Software that takes each inbound teaser and confidential information memorandum and gives the deal team a first-pass view of whether it fits the fund: the financials, leverage, add-backs, mandate and portfolio tests, sponsor record and red flags. Private Credit Deal Screening has six agents do that reading the day a memo arrives, and writes a one-page screen with every figure cited to its page.
The earnings quality analyst sorts each add-back into accept, haircut or reject under your written add-back policy. On Project Lantern it accepts $5.1M of one-time, incurred costs and rejects unactioned synergies and run-rate pricing, so leverage is 6.08× on the fund’s EBITDA rather than the sponsor’s 5.26×. An underwriter can change a treatment, with a reason that is kept.
Yes. Twelve tests run after the proposed hold: qualifying asset, sector, EBITDA range, leverage, add-backs, interest cover, sponsor equity, top customer, spread, single borrower, sector concentration and sponsor exposure. Each shows the value, the limit and its source.
No. The screen writer recommends and drafts; it never sends anything and never decides. The underwriting VP passes or recommends, the Head of Underwriting approves pursue, and an indication above $100M goes to the investment committee.
Yes. With past pass reasons on, a red margin note also shows what the fund decided the last time it saw the same problem, and what happened next. The Decisions page tracks passed deals after they closed elsewhere — on its sample data, 4 of 27 are now in trouble.
The intake reader never opens a data room without a signed confidentiality agreement, and the sponsor and comparables researcher never uses one borrower’s confidential information in another deal’s reply.
Yes. Leverage, add-backs, interest cover, EBITDA range, sponsor equity, top customer, spread, hold, sector and sponsor caps, excluded sectors, the add-back policy and approvers are all settings on the Mandate page. Changes are versioned and on the audit trail.
The Agentic Solution Engine builds and deploys it from your requirements — your mandate, add-back policy, screening template and a sample of past memoranda and decisions — and it goes live once every quality gate has passed. We will walk you through it on your own deals first.
Stressed loans flagged a quarter or more before non-accrual.
1-min demoPrivate Credit ValuationDocumented support for every mark, ready for committee and auditor.
1-min demoInvestor Reporting AutomationInvestor questions answered in a day, every figure traced to the books.
1-min demoWe’ll run Private Credit Deal Screening on a sample of your own memoranda, against your own mandate.