AnswersPrivate equity
In short
Operating partners need a platform that works across the portfolio, where each company builds on a shared foundation. Look for modules that are proven once and reused, acquisitions that join quickly, every frontier model so the plan never depends on one lab, security each management team can approve, and value measured against a baseline in every company. Fig builds this with CosmOS.
| Requirement | Why it counts across a portfolio | How Fig does it |
|---|---|---|
| One foundation for every company | Ten companies on ten different stacks means ten builds, ten security reviews and ten bills. | Each company gets its own CosmOS platform, built from the same foundation, so work moves between them. |
| Reusable modules | A workflow proven in one company should reach the next in weeks. | Proven modules are reused and adapted across the portfolio. |
| Acquisitions that join fast | Buy-and-build depends on integrating each add-on quickly. | Acquisitions and tuck-ins connect to the same foundation and the same digital twin. |
| Model choice over the hold | Model leaders change every few months, and a hold lasts years. | Every frontier lab and the best open-weight models, routed per task, plus models trained on the company's own data. |
| Security each company can approve | Every management team and its security lead must sign off. | SSO, SCIM, scoped roles, approval gates and a full audit trail, in the company's own cloud or Fig's. |
| Value measured company by company | The value creation plan needs figures the board can rely on. | Every system is measured monthly against a baseline set before the build. |
More of the return is built during the hold, through operating improvement the firm can measure. AI adds to it with shorter cycle times, more revenue captured by the same team, and work that grows without matching headcount. Most portfolio AI programs can make one company work; few make the tenth company cheaper than the first.
Scaling AI across a portfolio →Beyond cost: AI as a revenue engine →
The first workflows are usually the ones with volume, rules and a number the deal team already tracks: the monthly close and reporting, customer service and the front office, sales follow-up, procurement and scheduling. Fig is industry-agnostic, so the same platform serves a software company, a services business and an industrial company in the same fund.
Fig works with the firm and with each management team. A first deployment in one company puts a workflow into production in weeks and establishes the foundation. Proven modules are then reused and adapted in the next companies, acquisitions join the same foundation, and each company keeps its own data, systems and permissions. Every system reports against its baseline, so the firm sees value company by company.
The private equity AI playbook →Choosing an AI partner for your portfolio →
Start with one company and one workflow with a measurable number, put it into production in weeks, then reuse what it built in the next companies. A shared foundation makes each rollout faster and cheaper than the last.
Yes. Each company's platform runs in its own cloud account or Fig's, with its own permissions. Modules move between companies; each company's data stays with that company.
One with a workflow that has volume, clear rules and a number the deal team already tracks, and a management team ready to own it.
The platform is usage-based with no per-seat licenses, with budgets and spend rules for every team. Strategy, forward deployed engineering and training are scoped with the firm and each company.
Dream big, build fast, and grow far with Fig