About
Snyte answers plain-language questions against the sources you already run, and keeps the path it took so the number can be checked before it is trusted.
You ask a question in plain language. Snyte resolves it against the registered schema, shows the plan, runs it against your sources, and returns the number with the path it took. The generated SQL stays on the query record.
A Go API and a Next.js web application, shipped as containers with Kubernetes manifests and Compose files in the repository. There is no Snyte-operated cloud. You run it inside your own perimeter, next to the data.
AI features run through the shared EF gateway. Translation sends the schema; insight generation sends result rows. Local model serving is not available today.