Deploying
Deploying
Run agents-go in production — architecture, workers, scale, observe, drain.
How a single agent becomes a fleet: the two-plane architecture, running the agent as a worker, keeping latency low, observing what's happening, and scaling in without dropping live calls.
Architecture
Two planes — a worker fleet and a dispatcher — and the single-binary MVP.
Deploy a worker
Run the agent as a worker that registers, reports load, and drains.
Scaling & load
Load reporting, capacity, and how jobs are placed across the fleet.
Graceful drain
Draining a worker so a deploy or scale-in never drops a call.
Telemetry
Traces and metrics, and wiring an OpenTelemetry exporter.
Latency & interruption
What runs on the response path, and keeping barge-in responsive.