{
  "title": "Message Queues \u2014 Under the Hood",
  "description": "Emma and Ryan dig deep into message queue internals \u2014 from WAL-backed persistence and ack mechanisms to Kafka's partitioned log, RabbitMQ's exchange types, SQS visibility timeouts, and Redis Streams. They cover consumer groups, at-least-once vs exactly-once delivery, poison messages, dead letter queues, backpressure, ordering guarantees, observability with correlation IDs, the transactional outbox pattern, and when a queue is the wrong tool entirely. Starting with a real 3am war story about an 11-million-message RabbitMQ backlog caused by a single prefetch count change.",
  "topic": "message queues deep technical",
  "date": "2026-03-06",
  "episode_num": 6,
  "duration": "00:24:05",
  "size_bytes": 12881139
}