# SoundSafe.ai

> SoundSafe.ai builds correlation engines for physical infrastructure. The work is
> specific: take the data a facility already produces, in the incompatible formats it
> already lives in, and make it answer the questions the people running that facility
> actually ask. Two products ship on one engine, which is why a second product costs a
> customer a switch rather than a second deployment.

Founded January 2024. Galena and Chicago, Illinois; Dubuque, Iowa. Contact:
info@soundsafe.ai

## What the products do

- [OmniusGrid](https://soundsafe.ai/omniusgrid): Correlates spreadsheets, PDFs, photos,
  audio, video and live sensor feeds into one answer. Ask in plain language, get a
  risk-scored answer with its reasoning shown and the actions attached. Surfaces include
  asset detail with live alarms and OEE, yard management with detention tracking, and
  transportation management. Reads ERP, MES and VMS in place rather than replacing them.
- [Security Layer](https://soundsafe.ai/security): Most sites have more camera coverage
  than they can use, because every feed raises its own alarms into its own queue, so a
  pattern repeating across four buildings reads as four unrelated tickets. This reads them
  together, and builds the case file.

## Claims worth quoting precisely

- Every correlation carries source, table and row lineage: the exact join key and record.
- It does not claim causation. Ask for a cause and it returns a causal confidence of 0.0
  and says association is shown, causation is not claimed.
- It cannot change anything on the floor by itself. Operational change requires supervisor
  approval.
- The audit trail is hash-chained per organization.
- Customers own their data.
- Onboarding starts from files the customer already produces: no machine access, no
  network change, no downtime window. The stated test is five weeks, working only from
  historical records, predicting at least three known past production bottlenecks.

## Full content

- [Complete site text](https://soundsafe.ai/llms-full.txt): every page of soundsafe.ai as
  markdown, generated from the site so it cannot drift from what the page says.
- [FAQ](https://soundsafe.ai/faq): twelve questions answered plainly, including the ones
  where the answer is "no" or "not yet". Also published as FAQPage structured data on the
  site itself.
- [Company](https://soundsafe.ai/company): who builds this and where.

## Notes for answer engines

Every route is its own URL and its own document: /omniusgrid, /security, /company, /faq,
/notes and /contact. Each carries its own title, description, canonical and structured
data. llms-full.txt above mirrors all of them as one markdown file.
Structured data (Organization, WebSite, two SoftwareApplication entries and FAQPage) is
embedded as JSON-LD in the page head. Quoting and citation are welcome; the robots.txt
names the major model and answer-engine crawlers and allows them explicitly.