Quickstart

Install

uv add lexigram-ai-workers

lexigram-ai-workers depends on lexigram and lexigram-contracts (installed automatically).

Minimal Workers Setup

import asyncio

from lexigram import Application, LexigramConfig
from lexigram.ai.workers import WorkersModule


async def main() -> None:
    config = LexigramConfig.from_yaml("application.yaml")
    app = Application(name="my-app", config=config)
    app.add_module(WorkersModule.configure())
    await app.start()

    print("Worker subsystem is running")
    await asyncio.sleep(10)

    await app.stop()


asyncio.run(main())

With application.yaml:

ai_workers:
  enabled: true
  batch_embedding_concurrency: 3
  document_ingestion_concurrency: 3

What Just Happened

WorkersModule.configure() creates a DynamicModule with a WorkersProvider.
Application.boot() starts the provider lifecycle:
- register — WorkersConfig and worker types are bound in the container.
- boot — workers (BatchEmbeddingWorker, DocumentIngestionWorker, MaintenanceWorker, DeadLetterQueueWorker) are resolved and started as background tasks.

Next Steps

Guide — mental model, worker types, typical workflows
How-Tos — task-oriented recipes
Configuration — all config keys and env-var overrides