AI for private equity, built for the firm itself — not just the portfolio.
Fifteen to twenty inboxes of relationship context. Deal work scattered across SharePoint folders. Call transcripts locked in note-taker apps. A Dallas-based PE sponsor described the ask: a firmwide brain that actually answers “who knows someone at the target company?” We build and run that AI — inside your Microsoft 365 tenant, through your compliance review.
Official services partner of the platforms defining AI
NVIDIA
Anthropic
AI for private equity usually means the portfolio. The firm itself is still manual.
From five working sessions with a Dallas-based lower-middle-market PE sponsor — a partner wearing the value-creation and internal-AI hats, the COO as compliance gatekeeper: the firm’s knowledge lives in all sorts of different systems, and it is very hard to talk to all of them at once.
Knowledge in all sorts of systems
Relationship context sits in 15–20 inboxes, deal work in SharePoint project folders, call transcripts in AI note-taker apps. Getting an answer means knowing which system to ask — and who to forward the thread to.
“Our CRM is only as good as the data people put into it”
And people don’t put it in. There is a big gap between what’s in the deal CRM and what’s in the inbox — so the firm’s system of record is a partial record of the relationships it actually has.
“Does anyone here know someone at…?”
The highest-value question in the firm — who knows someone at the target, who the partners should be spending time with for sourcing — gets answered by whoever happens to be cc’d. AI should answer it from everything the firm has ever written down.
Banker outreach dies in the inbox
Every banker email is a potential deal — and a manual chore: read it, file it, key it into the CRM, pull the keynotes, decide who should look. Most weeks, triage loses to the calendar.
The AI bill is already a line item
The team is burning through frontier-model tokens — one person somehow hit $400 in a month and can’t say on what. Token usage really scales up on the deeper, more analytical questions — the ones worth asking.
Compliance can’t keep up with the models
As a registered investment advisor under Regulation S-P, every new vendor goes through vetting — and the compliance partners are way behind where the models are. “We all just use the enterprise subscription” is not an answer they accept.
Private equity deal sourcing runs on whoever happened to read the email.
What it looks like when relationships and deal flow stay manual at a ~15-person sponsor. Your numbers will differ — the pilot we run first puts figures on yours.
From anonymized sponsor-side calls — a Dallas-based lower-middle-market PE sponsor, ~15–20 FTE, an SEC-registered RIA across ~9 portfolio companies
Relationship intelligence, built from what the firm already wrote down.
We are an AI operating partner: we advise, build, and run the machine. Three steps, fixed order — your systems stay where they are.
- 01
Connect the systems the firm already runs
Read-only connections to Outlook, Teams, SharePoint, the deal CRM, and the note-taker apps — fifteen to twenty inboxes included. Nothing is re-platformed and nothing is re-typed; the AI reads the knowledge where it already lives.
- 02
Build the firm brain — with guardrails compliance signs
An ingestion and retrieval layer links people, companies, deals, and past theses into deep relational connections across your documents, notes, and messages. LP personal data stays out of scope by design; the SOC 2 report and due-diligence questionnaire reach your COO before the first sync.
- 03
Answer where the partners already ask
In Teams, email, or text: who knows someone at the target, what we’ve seen like this deal, what the last three calls concluded. Banker outreach gets filed, summarized, and scored against who the firm knows — risks, takeaways, action items, no re-typing.
Your private equity CRM is only as good as what people type into it.
The deal-CRM vendors will happily sell you a better empty database. But the buyer said it plainly: the CRM is only as good as the data people put into it manually — and partners don’t type. The gap between what’s in the CRM and what’s in the inbox is where the firm’s relationship intelligence actually lives.
So keep the CRM — even the legacy, slightly analog one. We build the AI layer that reads what nobody typed: the inboxes, the call transcripts, the SharePoint folders, the LinkedIn connections — then files, summarizes, and enriches the record automatically. The CRM stops being a typing chore and becomes a byproduct of work the firm already did.
We read news articles about stuff like this, but actually seeing it with our data, live, is a lot different.
Led by Peter Enestrom.
Asked by partners and COOs before the pilot.
Straight answers — including the compliance ones.
Not portco automation — the sponsor’s own four walls. AI reads the firm’s inboxes, call transcripts, SharePoint deal folders, and CRM, then answers the questions partners actually ask: who knows someone at this company, what have we seen like this deal, what did we conclude last time. The portfolio-side rollout — proving AI in one company and carrying the playbook across the companies you own — is a separate motion, covered on our PE-backed companies page.
An always-current map of who the firm knows and how — built from what people actually wrote and said, not what they remembered to type. Ask “does anyone here know someone at that company?” and the AI answers across fifteen to twenty inboxes, call transcripts, and LinkedIn connection data, with the thread to prove it. Identifying key relationships for prospective deals stops being a hallway question and becomes a search.
Banker outreach stops dying in the inbox. Each approach is read, filed into the CRM, and summarized — risks, takeaways, action items — then scored against who the firm knows and what it has done before. Screening a new deal means checking it against your own history: have we done anything specific here, and does it validate or invalidate the thesis? Partners spend their time on the deals the firm is positioned to win.
Through your gate, not around it. We run the vendor-vetting process your COO already applies: SOC 2 report, due-diligence questionnaire, and a deployment where everything — data, models, logs — stays inside your own Microsoft 365 and Azure tenant. LP personal data is excluded from scope by design. The compliance partners who are behind where the models are get documentation, not promises. Our on-premise AI page covers the deployment posture in detail.
The buyer’s own name for it was an enterprise, firmwide brain — one intelligent system that stores the firm’s data and answers across all of it, instead of a separate retrieval mechanism per system. Think of it as an associate that never forgets anything you’ve ever told it. It is not another platform to adopt; it is a layer over the systems you already run.
Because ad-hoc chatting sends every question — including the deep, analytical ones — to the most expensive frontier model. One person at the sponsor hit $400 in a month without knowing how. A firm brain fixes the economics as part of the build: an ingestion layer that doesn’t token-max the expensive model, and routing that shifts load to lower-cost models where answer quality allows it. We benchmark against firms your size and set the initial token budget with you.
Adoption fails when AI arrives as a new place to work. This answers where partners already ask — Teams, email, text — and starts with the questions they already ask out loud: who knows who, what have we seen like this. The comfort curve — when to ask AI versus when to do it yourself — comes from seeing it work on the firm’s own data, live, which is what the pilot is for. The buy-in question — why it’s worth investing, how it helps do more with less — gets answered with the firm’s own numbers, not a deck.
Start with one workflow.
Tell us where your team loses hours. We will come back with a straight answer on whether AI can help, what it would take, and what it would pay.