Head of Growth & Origination

TLDR:

We're hiring someone to build relationships across private equity and turn them into customers — funds, their operating and value-creation teams, and the portfolio companies they own.

In practice that means coffees, lunches, dinners, introductions. Today it's mostly me, at about a quarter of my time, and it's the main constraint on the business: our engineering team has capacity we can't fill fast enough.

If you want it, you can also run the engagements you bring in and work with customers directly — PE-backed portfolio companies with real problems and real data. It's the part most people find addictive. You'd see AI doing actual work inside an operating business: a year's worth of contracts becoming clean, cited data in Salesforce; a stalled M&A integration finally moving. Not a demo.

The people who do well here come from private equity or from inside a portfolio company. That's what earns the conversation — you've sat on a deal team, run an integration, or owned a function when the data was a mess, so you can talk about the problem the way the buyer actually experiences it.

What we're building

We're the data transformation partner for PE-backed portcos. We turn the documents a business is buried in into clean, trusted data in the systems it runs on — and, harder, we get the organization to actually adopt it.

The work we get called in for:

• M&A data migration. A roll-up has bought a dozen companies and none of the data ties out. Contracts, customers, and billing live in whatever system each acquisition came with.
• ERP and CRM migrations. The move to NetSuite or Salesforce that's been stuck for a year because nobody can get the legacy data clean enough to load.
• Rev ops and contract data. Pricing, terms, renewal dates, and entitlements buried in signed PDFs — so invoicing is wrong, revenue leaks, and nobody trusts the pipeline.
• Back-office and finance operations. Invoices, statements, and filings that a team of people still retype into a system of record every month.
• Failed implementations. The systems integrator took eighteen months and it didn't work. We get called to fix it.

How we do it:

• Getting the data out is the entry ticket. Contracts, invoices, statements, credit agreements, rent rolls — scans and handwriting included, at production accuracy, with every value citing its source page. Most vendors can't do this part. It's necessary, and it's not sufficient.
• The hard part is everything around it. These projects don't fail on model accuracy; they fail inside the company. Whose numbers change when the data is finally right. Which VP has to stop running the business out of their own spreadsheet. Who owns the field in Salesforce, who signs off on the migration, who tells the sales team their commissions were wrong. That's change management, and it's most of the job — which is why we do it in the building, with the people, rather than over email.
• Auditability is what makes it survivable. Page number and bounding box on every value, full audit trail. It's what lets a CFO trust AI-extracted data enough to bill off it — and it's the thing customers tell us actually sold them.
• We land it in the systems the business runs on. Salesforce, NetSuite, Snowflake, Databricks. The work isn't done when the model is accurate; it's done when the data is live and people are using it.
• Then we build the layer above it — the software, automations, and agents a company needs to run better and be worth more at exit.
• We embed. Our engineers go on-site and into the customer's Slack. We scope the real problem in about a week and own the outcome to production. Four to eight weeks — not a nine-month statement of work.
• The platform compounds. The same problems recur across companies, so every deployment makes the next one faster and more accurate. It's the one thing a general-purpose model can't copy — it's built out of real customer data and the review that comes with it.

What that's looked like in practice:

• A sponsor-backed software company had a multi-million-dollar systems-integration project fail, leaving hundreds of legacy contracts stranded and its billing wrong. We rebuilt it in weeks: contracts into Salesforce at ~99.5% field accuracy, every value cited.
• A services roll-up had acquired twenty-plus companies and couldn't tell you what it was owed. We reconciled the acquisition data into a single source of truth and surfaced the revenue it was leaking.
• A regulated financial institution's filing process — two people, by hand, across eighteen entities — now runs in about twenty minutes, with every figure traceable to its source page.

Our customers

• Private equity funds — over half of the top 20 funds in the world already work with us. The relationship starts with the fund, and the fund is what gets us in the door everywhere else.
• Their portfolio companies — where the work happens, and where the money is. The sweet spot is $50–300M revenue, because the sponsor can actually mandate the fix.
• The people who sign: operating partners and value-creation teams, portfolio-company CFOs and CTOs, and the deal teams who introduce us.

One sponsor owns hundreds of companies with the same problem. One win is how we get the next twenty.

Why now

• Every fund has an AI mandate and almost none of them have a way to execute it inside the portfolio. The deal team using AI is table stakes; the portfolio actually running on it is the return.
• The incumbents are the big consultancies and the systems integrators. They're expensive, slow, and their projects fail often enough that “we tried this already” is a common opening line.
• The AI labs don't do this work outside of large cap PE’s largest portcos — their minimum engagement is $5M+, and they're not going to sit in a portfolio company's Slack for six weeks.
• So the field is open, and almost nobody credible is talking to private equity about it.

The job, in three parts

1. Relationships — the core of it. Build and work a network across private equity: operating and value-creation teams, EIRs, deal teams, and the CFOs and COOs of the companies they own. Catch up with the people you already know, get introduced to the ones you don't, and go to the cities where they are. Turn those conversations into customers — a portfolio company with dirty contract data, a stalled M&A integration, a vendor implementation that failed, an ERP that doesn't reconcile.

2. Engagements — optional, and encouraged. Run the work you bring in as the engagement manager: sit with the customer, scope the problem, direct our engineers, own the outcome. You don't have to, and the commission reflects the choice. It's also the fastest way to get better at part one — being able to describe exactly what broke at the last portfolio company, and how we fixed it, is what makes the next conversation land.

3. Building the practice — over time. AI and data transformation inside the portfolio is one of the largest unmet needs in private equity, and there's very little credible, first-hand writing on it. Publish, speak, host dinners, help us start a podcast. If you want to build out a line of business — the CFO practice, say — we'll hire under you and you'll own that book.

Who you are

You spent time in private equity. A deal team, an operating or value-creation team, or a function you ran inside a portfolio company. That's what earns the conversation — you can talk about the problem the way the buyer experiences it, not the way a vendor pitches it.

You have a deep network across private equity. People who reply to your texts / are your friends. Ideally it runs across:

• Mid-market funds, where most of the opportunity is and almost nobody has been sold to yet.
• The larger funds, where you came up.
• Portfolio-company executives — the CFOs, COOs and CIOs who actually sign the work.

And the rest:

• You like the work of building relationships. Coffees, dinners, introductions — this is the job, not the overhead around it.
• You've seen the mess first-hand. Post-acquisition data chaos, rev ops that doesn't tie to the GL, four consulting firms and it's still manual.
• You're genuinely interested in AI and want to see it do real work — inside an operating business, on real data, not in a demo.
• You want the upside. Compensation here is weighted toward what you build.
• You're careful with your name. You only get to introduce your network to one thing. We understand that, which is why we'd rather show you the work than pitch you.
• You're at the point where you want out. You've seen what the next eight years look like if you stay, and you'd rather own something.