⚡AgentSkills
⚙️ Backend Engineering · Caching & Queues

Build trustworthy background job processing

Queue async work with retries, dead-letters, and observability so failures surface instead of vanishing.

intermediate~35 minBackend EngineersAPI DevelopersPlatform Engineers

Steps

  1. 1Make handlers idempotent; assume at-least-once delivery
  2. 2Set exponential backoff with jitter; cap retries at sane counts
  3. 3Dead-letter poison messages with full context for triage
  4. 4Separate queues by latency class and priority; prevent head-of-line blocking
  5. 5Emit metrics: depth, age of oldest job, success rate, duration percentiles
  6. 6Alert on oldest-job-age, not just queue depth

Common Pitfalls

  • ▲Long jobs blocking a shared queue of quick tasks
  • ▲Silent retries masking a permanently failing handler

Commands

Install with skills CLI
$ npx skills add aniruddhaadak80/skills --skill caching-queues-background-jobs
Install globally
$ npx skills add aniruddhaadak80/skills --skill caching-queues-background-jobs -g

Tags

#queues#async#reliability#backend-engineering#caching-queues

Related skills

Decide what caches where (CDN, app, Redis) and how entries die so staleness stays bounded.

⚙️ Backend Engineering·~35m