Hatchet Has a Free Task Queue — Durable Workflows for Background Processing

#backend #queue #typescript #go

Hatchet is an open-source distributed task queue with durable workflow execution.

What You Get for Free

Durable workflows — survive crashes, auto-resume
DAG workflows — define step dependencies
Concurrency control — limit parallel executions per workflow
Rate limiting — control throughput per workflow type
Retries — configurable retry strategies
Cron scheduling — recurring workflow triggers
Events — trigger workflows from events
Dashboard — visualize running workflows
SDKs — TypeScript, Python, Go
Self-hosted — free, unlimited workers

Quick Start

# Docker Compose
git clone https://github.com/hatchet-dev/hatchet.git
cd hatchet && docker compose up -d

import Hatchet from '@hatchet-dev/typescript-sdk';

const hatchet = Hatchet.init();

const workflow = hatchet.workflow({
  name: 'order-processing',
  steps: [
    hatchet.step('validate', async (ctx) => {
      return { valid: true };
    }),
    hatchet.step('charge', async (ctx) => {
      // Durable — survives crashes
      await chargeCard(ctx.input.orderId);
    }),
    hatchet.step('notify', async (ctx) => {
      await sendEmail(ctx.input.email);
    })
  ]
});

Why Teams Choose It Over Celery/Bull

Celery loses tasks on crashes. Bull needs Redis:

Durable — workflow state persists through crashes
DAG support — complex step dependencies
Dashboard — see every workflow execution
Multi-language — TypeScript, Python, Go SDKs

A payment processor lost 3 orders during a server restart — Celery tasks vanished. After switching to Hatchet, workflows resume exactly where they left off after any crash. Zero lost transactions in 6 months.