@kodekloud: System Design: Scaling Webhooks to 10 Million Events/Day! #shorts A full system design walkthrough: how to build a webhook delivery system that reliably sends 10 million events a day. We go from a naive POST that drops events, to a durable, fault-tolerant pipeline: Transactional Outbox → fair-share dispatcher → retries with exponential backoff → HMAC signing → Dead Letter Queue + replay. One slow customer can take down your whole platform — here's how to make sure it never does. 🚀 Subscribe for the system design answers that actually land offers. 💬 How would you handle the slow-customer problem? Comment below. #systemdesign #webhooks #softwareengineering #backend #distributedsystems #coding #programming #softwarearchitecture #systemdesigninterview #scalability #tech #backenddevelopment #microservices #faangprep #engineering #codingshorts #softwaredeveloper
I mean, it’s simpler to just post to an R/BMQ queue and let FIFO workers handle appropriately. If the server fails to send, the worker re-posts it to the queue and increments a try counter. After N attempts its send to a dead letter queue for later reprocessing. More specifically, the semantics with RMQ is producer -> quorum queue -> manual acks + delivery limit + DLX.
2026-06-05 03:06:20
1
interdurozw :
115 events per second is nothing. My engine handles 10000 events per second in my home PC
2026-06-05 03:02:41
0
Dn :
The replay UI is British
2026-06-05 02:18:29
0
Thamie :
Scaling is a new level when it comes to software engineering
2026-06-05 03:01:21
0
2af :
🥰🥰🥰
2026-06-04 19:29:41
0
Hammad Elaraby :
💥💥💥
2026-06-05 12:15:49
0
To see more videos from user @kodekloud, please go to the Tikwm
homepage.