Skip to main content
Smaya

​
Purpose

Smaya is an enterprise Agent OS: one governed environment in which an organization builds, deploys, monitors, evaluates and continuously improves a digital workforce. You do not start with an empty canvas. Smaya ships a library of pre-built agents — specialized workers, each accountable for one job — and a way to combine them into AI Employees: named roles with a defined mandate, bounded authority, an approval model and a named human owner.

​
The agent swarm

Upcoming. Give Smaya the role. It builds the pipeline, talks to every candidate, gathers the evidence, and brings you the people worth deciding on. That is an agent swarm: one run per role, one sub-run per candidate, built from pre-built agents. You will answer reports in the Agent Inbox and never manage the agents themselves. Available today: AI Employees in chat, Smaya ask and report chat, the MCP server, and screening through to a ranked Leaderboard. Upcoming: starting a role run, the Agent Inbox and the full role-to-decision run. See The Agent Swarm.
Land with Exterview. Expand with Smaya. Most organizations meet Smaya through one concrete problem — usually hiring. The platform underneath is the same one every later workload runs on, which is why the second application costs a fraction of the first.

​
The four layers

The OS provides governance, integration and lifecycle. Applications carry the domain expertise. The Agent Library provides the pre-built workers. AI Employees are what your people actually work with.

1. The Agent OS

Governance, integration, orchestration, evidence and lifecycle management. Built once, shared by every application above it.

2. Applications

Deep domain expertise for one business function. Exterview is the first, covering talent. Gradientflo is next, covering everything after the hire.

3. Agent Library

Pre-built, specialized agents. 26 in Exterview, ten Upcoming for Gradientflo, on a path to 100 across the platform.

4. AI Employees

What your teams work with: named roles such as Role Designer, Screener, Coordinator and Report Specialist, each doing one kind of work. See AI Employees.

​
Why an OS rather than a set of tools

Every organization adopting AI at scale hits the same three walls: nobody can say what the AI decided and why, nobody can bound what it is allowed to touch, and every new use case is built from scratch.

Evidence by default

Every consequential decision produces a permanent record: what was assessed, against which criteria, on what evidence, reviewed by whom. Not a log file — a business artifact.

Authority is bounded

An agent never exceeds the authority of the person who invoked it, and never exceeds its own declared scope. Anything consequential waits for a human.

Built once, reused

Identity, permissions, approvals, audit, integrations and evaluation are platform services. A new workload inherits them instead of rebuilding them.

​
Applications on Smaya

See Applications for how the boundary between them is drawn.

​
Built on Microsoft

Smaya runs on Microsoft Azure and grounds itself in the customer’s own Microsoft 365 environment — identity, calendar, meetings and mail — through Microsoft Graph, and aligns with Microsoft Work IQ as an enterprise context layer. See Microsoft & Interoperability.

​
FAQs

No. Smaya is the platform. It includes a conversational assistant as one surface, but the substance is the governed environment in which AI Employees are configured, deployed and supervised.
No. Organizations typically start with one application solving one concrete problem. The platform capabilities come with it, and additional workloads activate on the same foundation.
Exterview is live and transactable on the Microsoft Marketplace. Each of the 26 Exterview agents is labelled Available or Upcoming on Exterview Agents. Gradientflo and its ten agents are Upcoming. Every page states the status of what it describes.
The platform direction is a growing library of pre-built agents that customers configure, rather than a blank development environment. Scoring criteria, thresholds, tone and the agent’s role for each job are yours to configure today. Composing your own AI Employees from the library is Upcoming.