The challenge
Infinity Constellation operates eight portfolio companies across different verticals, each with its own CRM, revenue-management platform, and business definitions. There was no shared language for core concepts like leads and proposals. Leadership decisions waited on manual reports that were stale by the time they landed, and portfolio companies burned hours on reporting overhead instead of business development.
What we built
Paradox Machines built a centralized cloud data warehouse spanning all eight portfolio companies, deployed inside Infinity’s own cloud infrastructure. The implementation included:
- Standardized business logic and definitions across every portfolio company
- API integrations connecting each company’s CRM and revenue systems
- Clean, modeled data exposed to Claude via a custom MCP connector
- A natural-language query interface that meets stakeholders where they already work
Rather than deploying another portal, we made Claude the interface — leadership asks a question and gets a live, trustworthy answer, instead of filing a request and waiting on an analyst.
How we did it
The engagement prioritized lightweight, production-focused execution:
- 1
Requirements gathering
~30 minutes per portfolio company to map core object definitions
- 2
Deployment
a few days of configuration within Infinity's cloud environment
- 3
Integration
minutes per business unit once API keys were provided
- 4
Data modeling
standardized business logic across eight companies into one shared schema
- 5
Claude connectivity
a few clicks to configure the MCP connector
Once complete, data flowed into the centralized warehouse in real time.
The role of AI
Claude sits on top of the centralized, standardized warehouse via the MCP connector. Non-technical leadership asks it questions directly — surfacing opportunities, tracking KPIs, and pulling reports without an analyst in the loop or a wait for data to be verified.
The results
- Same-day decisions, not next-week reports — leadership queries live portfolio data directly in Claude instead of waiting on an analyst
- Eight companies, one shared view — every portfolio company now speaks the same data language
- $100K in annual savings from eliminating manual reporting, plus ~2 hours saved each week per company
Best fit
Holding companies, private equity firms, and multi-entity operators who need leadership to act on portfolio data the day it changes — not whenever an analyst can turn around a request.