Jun 13Vibe with Hermes Agent — Bengaluru · RSVP
ToolsMCPBlogResearchCommunityStar on GitHub
All Tools
B
Dev ToolsFreemiumOpen Source

BULLMQ

Redis job queues for background AI workloads

MIT

ABOUT

LLM calls, embedding batches, and agent runs cannot live in the HTTP request path without timeouts and lost work. BullMQ puts those jobs on Redis (or PostgreSQL) with retries, backoff, concurrency, and delayed schedules so workers process them independently of the API process and failed jobs can be inspected and replayed instead of silently dropped.

INSTALL
npm install bullmq

INTEGRATION GUIDE

1. Queue embedding generation and document-ingestion jobs so web requests stay fast 2. Retry failed LLM API calls with exponential backoff without blocking the user request 3. Run nightly eval batches and agent workflows as scheduled workers with concurrency limits

TAGS

nodejsredisqueuejobsworkersbackground-tasks