Use cases

What you build with Golem

Golem is the durable agent runtime. From per-user AI agents to multi-step business workflows, here's what it's built to run — and the architectural reasons why.

Classic durable execution

Not just agents — any workflow that has to survive

Overview

The workflows that have to finish — even through restarts, retries, and outages. Multi-step workflows that span days, payment flows that have to either fully complete or fully roll back, scheduled jobs that have to run on time even through crashes. It predates the agent era, and on Golem, the same runtime that runs your agents runs it just as well.

Examples

  • Charge a card, update the ledger, send the receipt — all-or-nothing, even through crashes
  • Subscription billing engines with retries, dunning, and proration
  • Cron-driven pipelines replacing Celery, BullMQ, Sidekiq, Step Functions
  • Webhook reconciliation across systems with different ordering semantics

Why Golem

  • Durability is the primitive — no opt-in, no manual checkpoints
  • No effect ever fires twice — retries don't double-charge
  • Long-running natively — multi-day workflows, no serverless timeout
  • Same runtime as your agents — one cluster, one operations surface

Crash your first agent in five minutes.
Watch it come back.

Scaffold a durable agent, run it locally, kill the process at any line, and watch it resume exactly where it stopped.

# Install: download from github.com/golemcloud/golem/releases
golem new --template ts --component-name example:counter --yes my-agent
cd my-agent && golem build
golem repl