PPRR × Carnegie
Proof of delivery

Every one of your seven ideas maps to something PRR already runs in production.

We're not proposing research projects. Intelligent document processing, multi-source knowledge graphs, autonomous monitoring, agent orchestration, secure role-based access — each is a system we've shipped. Here they are, mapped to your use cases.

🧾

KreditIQ

Intelligent document processing · production SaaS

AI commercial-loan underwriting: ingests financial documents, extracts and analyzes data, scores risk, and generates credit memos with a human approval gate.

Proves 1 · Concur receiptsProves 3 · Data extraction
Same OCR → extract → structure → human-review pipeline the Concur agent uses.
🔎

Marsh Prospector

Enrichment + AI drafting · live at marsh.theprrgroup.com

Maps 25,000+ entities from public records, enriches each with identities and contact data via connectors, and generates Claude call strategy + email drafts.

Proves 3 · Contact / CRMProves 2 · Research + summarize
Exactly the enrich-dedupe-summarize loop behind the CRM and library agents.
🗺️

TIPSTR (USF Risk-to-Resilience Lab)

Multi-source KMS + entity graph + dashboards

Ingests signal data from many federal/state/county sources, normalizes and triages it with LLM extraction, builds a cross-source entity graph, and serves drill-through dashboards and one-click reports.

Proves 4 · China AI KMSProves 2 · Library research
The blueprint for the China AI tracker and the Climate DDD dataset — down to the map + drill-through.
📡

Deal Signal

Autonomous monitoring + scoring engine

Continuously scans public sources for signals, matches them to services, scores 0–100, and surfaces the decision-maker — an always-on background agent.

Proves 5 · QC monitoringProves 4 · Background agents
Proves the scheduled-scan-and-flag pattern behind the QC monitor and KMS updater.
🧠

Triage-OS

Autonomous multi-agent orchestration

Ingests email, Slack, Notion, calendar, and GitHub signals, turns them into prioritized tasks against live projects, and executes — with humans in the loop.

Proves 6 · Event logisticsProves 7 · Ask-Andy
The orchestration + retrieval layer under the event copilot and the intranet agent.
🧭

Clarity

Agentic platform · secure microservices

A production agentic case-management platform with role-based access, audit trails, and deploy-anywhere architecture.

Proves PlatformProves Security & governance
Shows we run agents securely at organizational scale — not just prototypes.

The research-institution reference you asked for

For a peer reference, we can connect you with the University of South Florida (an R1 research institution) — including Dr. Shelly Wagers, who leads a statewide lab, and our partners at Innate Health Research. TIPSTR, shown above, is the platform we built with USF's Risk-to-Resilience Lab.

Organizations PRR supports

Homeland Security InvestigationsU.S. Dept. of Health & Human ServicesU.S. Dept. of JusticeUniversity of South FloridaMarsh McLennanPenguin SolutionsEdgeConneXLos Angeles RamsStarkeyLeprino FoodsUniversity of North CarolinaOrlando Int'l AirportJacksonville JaguarsKreditIQAmwinsNFP
See the platform architecture