Features deep dive

Four building blocks, one platform. Every layer is replaceable.

Bring your own models, your own sources, and your own servers — or take what ships and tune it. Below: what each block actually gives you.

01 · Flow orchestration

The assistant doesn't just reply — it finishes the job.

Build a flow that runs steps in order, branches on conditions, and calls your tools. Start the same flow from a conversation, a schedule, or an event.

  • Build flows from drag-and-drop steps
  • Conditional branching and loops
  • Scheduled and event-driven tasks
  • A step-by-step record of every run
02 · Answers from your documents

See where every answer came from.

Collect documents, tables, web pages, and connected sources into one knowledge pool. The assistant builds its answer from that pool and shows the passage it used underneath.

  • PDF, Word, Excel, web, and connected sources
  • A source link under every answer
  • Update the source and the answer follows
  • Quality runs that measure accuracy
03 · Model governance

You decide which job runs on which model.

Put cloud providers and models on your own servers into a single list; per-team permissions, usage limits, and cost stay on the same screen.

  • One provider-agnostic model list
  • Per-team and per-user model permissions
  • Usage limits and credit tracking
  • Swap a model without rebuilding anything
04 · Enterprise architecture

Your data stays wherever you decide.

Teams, roles, and fine-grained permissions; secrets encrypted at rest, an audit trail, and single sign-on. Run it in our cloud or entirely on your own servers.

  • Teams, roles, and fine-grained permissions
  • Secrets encrypted at rest
  • An audit trail for every change
  • Cloud, self-hosted, or a fully offline install

See all four against your own data.

Upload a few documents, pick a channel, and test it with real questions. We'll go through what changed together.