One standard for background jobs. Every language. Every backend.

Enqueue in JavaScript, process in Go

OJS jobs are plain JSON. Any language can produce them, any language can consume them. The server handles retry logic, scheduling, and state management so your clients stay thin.

// JavaScript: enqueue a job
import { OJSClient } from '@openjobspec/sdk';

const client = new OJSClient({ url: 'http://localhost:8080' });
await client.enqueue('email.send', ['user@example.com', 'welcome']);

// Go: process the same job
worker := ojs.NewWorker(ojs.WorkerConfig{
    URL:    "http://localhost:8080",
    Queues: []string{"default"},
})

worker.Handle("email.send", func(ctx ojs.JobContext) error {
    to := ctx.Args[0].(string)
    template := ctx.Args[1].(string)
    return sendEmail(to, template)
})

worker.Start(context.Background())

---

Why OJS?

Backend-agnostic

Write your job logic once, run it on Redis, PostgreSQL, or any conforming backend. Switch backends without changing application code.

Language-agnostic

A Python service can enqueue a job that a Go worker processes. Six official SDKs cover the most popular languages, and the spec is simple enough to implement in a weekend.

Battle-tested design

OJS synthesizes the best ideas from Sidekiq, Faktory, Oban, BullMQ, Celery, and Temporal. Every design choice is documented with rationale and prior art.

---

SDKs for every language

Go

go get github.com/openjobspec/ojs-go-sdk

Full client + worker, middleware, workflows

GitHub →

JavaScript / TypeScript

pnpm add @openjobspec/sdk

ESM, vitest tested, full type definitions

GitHub →

Python

pip install openjobspec

asyncio, mypy strict, Python 3.11+

GitHub →

Java

Maven: com.openjobspec:ojs-sdk

Java 21 records, zero required dependencies

GitHub →

Rust

cargo add openjobspec

tokio async, serde, full type safety

GitHub →

Ruby

gem 'openjobspec'

stdlib only, Ruby 3.2+

GitHub →

---

Choose your backend

Both official backends support all five conformance levels and are production-ready.

Feature	Redis Backend	PostgreSQL Backend
Storage Engine	Redis 7.0+	PostgreSQL 15+
Atomicity	Lua scripts	SQL transactions
Dequeue Strategy	ZPOPMIN / sorted sets	SELECT FOR UPDATE SKIP LOCKED
Persistence	RDB + AOF (configurable)	WAL (always durable)
Horizontal Scaling	Redis Cluster	Read replicas, Citus
Real-time Notifications	Pub/Sub	LISTEN/NOTIFY
Conformance Level	Level 4	Level 4
Best For	High throughput, low latency	Durability, existing Postgres infra

---

Get running in 5 minutes

One Docker Compose command, one SDK install, and you have a working job system.

Quickstart GuideRead the Spec