If a request waits for a PDF, email, image resize and three webhooks, the user is paying for work they do not need to watch.
Move slow and retryable work to a queue.
But a queue needs more than <code>add()</code>:
<ul>
<li>Idempotent processors</li>
<li>Retries and backoff</li>
<li>Dead-letter handling</li>
<li>Job-status visibility</li>
<li>Monitoring</li>
<li>Safe payloads</li>
</ul>
A queue should make failure manageable, not move failure somewhere nobody checks.
What produced the biggest improvement after you moved it to a queue?
#NestJS #BullMQ #BackgroundJobs

Welcome to the NestJS Developers Hub, a collaborative community for developers passionate about NestJS! Whether you're a seasoned expert or just getting started, this group is dedicated to sharing knowledge, solving problems, and advancing our skills

Nestjs Developers

Offloading slow, retryable work like PDF generation, emails, image resizing, and webhooks to a background queue keeps requests fast. Building a reliable queue requires more than just adding jobs: idempotent processors, retries with backoff, dead-letter handling, job-status visibility, monitoring, and safe payloads are needed so failures are handled rather than silently lost.

Day 17/30 — Queue the Slow Stuff

Yes Absolutely. This principle we have used is you API process should only be serving API. It should not be engaged in any process that is not directly impacting the user in realtime. Move those as workers that will run asynchronously based on a message. This message can be an explicit call from API which goes to a message queue , or based on a change raise an event that triggers this process from one of the microservice. Initially we built a task platform inhouse. Later we decided to move this to AWS serverless purely to reduce cost and avoid scaling maintenance. Now that said, you still can have small micro back end task with in API. But it should not span for even couple of seconds.

The biggest improvement was reducing server load. Moving the work to a queue helped offload long-running tasks from the main request path, which also reduced failures during long tasks.

Great post! Thanks for sharing these insights on managing background jobs 🚀