Single Source of Truth
Transform messy private asset reporting into clean, traceable data.
Made possible through our AI-Native Data Pipeline
AI-Native Data Pipeline
How does this all work?
Document retrieval
We collect holdings reports, capital account statements and cash flows from your administrators and aggregators, or retrieve them from the systems where you hold them.
Extraction and ingestion
Each figure is read from the document it arrives in, keeping its provenance, and checked against the statement's own totals and the fund's earlier reports. Anything that does not add up goes to an analyst, who corrects it at its place on the page.
Cleaning, enrichment and normalization
We fix extraction errors, replace restated reports, and standardize units, currencies, dates and fund names. Then we add the metadata needed for agentic analytics.
Reconciliation to the source
Data is verified across independent sources, with discrepancies flagged and confirmed directly with the originator.
The data store
We record every value into the Valuation Book of Record, then use a knowledge graph to link each fund to its GP, portfolio companies, key people and news for a total portfolio view.
The data intelligence layer
Reach it however you work, in our apps, in Excel, through the agentic interface, or from your own pipelines and agents through our API.
Data Extraction
Rune turns PCAPs, schedules of investment and company fundamentals into clean usable data. Every number traces back to the exact page and line it was read from, so anyone can check it.
PDFs to Data
Every figure traced to its line on the page.
GP reports arrive as PDFs, spreadsheets and Word files, each in its own format, and most of the data in them is unstructured. We read every figure from the page it arrives on, with its location, so it can be checked back to its source.
Valuation as code. Not AI.
Fully automated, proprietary deterministic models value every holding every trading day. The valuation is computed by mathematics rather than a language model, so the same inputs return the same value every time, and each value traces back to the data behind it.
We exercise the data, every day.
Each estimate draws on thousands of data points, not a simple ratio or beta: the schedule of investments at the holdings level, asset-class-specific data and signal, and fundamentals. The same data feeds our alerts and monitoring.
Data transfer
- 01
Retrieval, run by us
We collect your statements wherever they arrive today, in a shared drive, a dedicated mailbox or a team channel, in whatever format they come, and check for new ones on a schedule we agree with you.
- 02
API
Your systems, or a provider that already extracts your statements, send data directly into ours.
- 03
SFTP, set up by us
For files, we set up an SFTP site and run it on your behalf.
Your data is protected by audited controls.
Our controls are covered by SOC 1 Type II and SOC 2 Type II reports. SOC 1 covers the controls that bear on our clients' financial reporting, and SOC 2 covers security, availability, processing integrity, confidentiality and privacy.
Each report comes from an independent auditor who tested our controls over a period of time, not on a single day.
All client data is processed and stored in the US and never leaves the United States.
Breakthrough technology. Disruptive pricing.
The most direct test of a valuation method is to watch it run on a real portfolio.
We will value one of your portfolios.