Sandro Dezerio
Senior fullstack engineer, 9+ years. Node.js, TypeScript, AWS.
I write about reliable AI workflows and event-driven systems, with small runnable examples.

Series · 10 parts
Reliable background jobs in Node.js, from zero
Ten parts on background jobs in Node.js. Each one adds a single idea to a small queue you can run on your laptop.
Parts 1 to 7 are live.
What is a job queue, and why does my API need one?
How many jobs should a Node.js queue run at once?
How to retry failed jobs in Node.js without making it worse
Ten steps at 99% is 90%: retrying multi-step jobs
Jobs that run twice: idempotency keys in Node.js
Jobs that survive a crash: Node.js and BullMQ
Failed jobs in Node.js: dead-letter handling with BullMQ
Latest posts
-
Failed jobs in Node.js: dead-letter handling with BullMQ
BullMQ has no built-in dead-letter queue. Stop retrying permanent errors, park failed jobs, alert on depth and age, and replay safely.
-
Jobs that survive a crash: Node.js and BullMQ
A queue in memory loses every waiting job when the process stops. See how BullMQ and Redis keep them, tested by killing the worker.
-
Jobs that run twice: idempotency keys in Node.js
A retry can run a side effect twice. A simulation of 1,000,000 sign-ups shows how idempotency keys stop duplicate emails, and where the key…