Product primitive
Realtime Tasks
Coordinate live work between humans and AI agents. A Task makes progress and participation visible without moving the agent or your source of truth into Lynta.
A Task is a live unit of work
Agent work can continue after the request that started it has returned. A Task gives an application a unit to follow while that work progresses, pauses for input, or completes.
A Task can expose status, progress, steps, questions, responses, completion, failure, or cancellation to the application experience. Lynta coordinates realtime information around the task; your service still runs and controls the work.
Subjects identify the participants
A Subject identifies a participant in a Task. It can represent a person, an agent, or a service. Subjects belong to the Task model; they are not a separate top-level primitive.
user:123— a person using the application.agent:researcher— an agent performing part of the work.agent:reviewer— another agent reviewing or extending the work.service:payments— a service participating in the workflow.
Build human-in-the-loop experiences
For example, a research agent can show that it is analyzing sources, ask which market to prioritize, and wait for a person to answer. The application can show the question and response as part of the live Task, then the agent service can continue its own work.
More than one human or agent can participate in a Task. This makes it possible to present a researcher, a reviewer, and a human operator together in the same product workflow.
Lynta coordinates the realtime layer
Lynta does not execute an AI agent, choose its model, or replace your orchestration system. Your own backend remains authoritative and runs the workload. Lynta makes task-related changes and events live to connected participants.
Use State for backend changes that should sync to clients and Channels for ephemeral bidirectional events.