Skip to content
Optera Pre-launch · 2026
Get access

The source registry

Everything goes in. This is everything.

The engine's intake, itemized across four domains: financial, physical, civic, digital. Every feed, who operates it, what we draw from it, and how often it lands. If a number shows up in a product, it started somewhere on this page.

Measured, not rounded up

60+
live data sources[1]
5.9M
data points ingested / day[2]
189
data pipelines[1]
77M
records under management[3]

[1] the engine's sync registry · [2] ingest log, trailing 93 days · [3] live record count · figures as of 2026-07-19. Rows below group related feeds, so this page reads shorter than 189; the registry is the count of record.

The intake, by domain

Skip the registry

Domain 01

Financial

What a company reported, what the economy did, and what the market paid for it. Three groups, and every figure in them ends at the document it came from.

01

Filings & disclosures

The primary record: what companies file, read at the source.

02

Macro

The economy's own numbers, from the agencies that publish them.

03

Markets

Prices, flow, and the paper trail around them.

Domain 02

Physical

Where the money actually is: hulls, weather, aircraft, fire and ground truth, read on the same clock as the markets that react to them.

04

The physical world

Ships, storms, fires, and rockets. Portfolios live in the world, so the engine watches it.

Domain 03

Civic

What the state does to a position before it is priced in, and what it has to disclose while doing it.

05

Washington

What the government trades, buys, approves, and investigates.

Domain 04

Digital

The traces a business leaves while it is still too early to show up in a filing.

06

The internet

Attention, hiring, shipped code, ad spend: the exhaust of doing business.

Optera engine

Every row on this page lands in the same machine. Eight beats later, it leaves as intelligence.

Run the engine →

Below the intake

Made in the engine.

These are not sources. They are what the engine computes from the rows above, and they are labeled as computed everywhere they appear.

Scores & gradesConvexity scores, factor grades, conviction and analyst scorecards, recomputed daily.
Valuation baselinesDCF baselines and forecasts, rebuilt from filings and prices, with their assumptions attached.
The exposure graphEntities, relationships, event studies, and revenue-at-risk scenarios (Mimir).
Search & retrievalEmbedding indexes over filings, transcripts, and releases, so questions find documents.
AI enrichmentSummaries and event tags, always carrying the source they were read from.

Operators are named where the record is public, and credited where our licenses allow; the rest are described by class. Cadences are the engine's scheduled sweeps, not the upstream's publication rate. Feeds come and go as coverage expands; the registry is refreshed with each release. Your linked accounts in Vexa are your own data, never part of the shared substrate. Registry as of 2026-07-20.

Get access

Four domains. One question bar.

Optera is pre-launch. Join the early list and we will bring you in as each surface opens.

Request access