A post about the cache stampede problem. When a hot post goes viral and a million people ask for the same data at the same second, even a Redis cache in front of the database is not enough. This walks through single flight, the Redis lock, and using pub/sub so all the waiting requests get the result the moment the first one finishes.

A practical walk through of Kafka, first the entities in plain words, then a real high scale use case (an uptime monitoring service checking 10 million endpoints), and finally how to actually pick your partitions, consumers, brokers and consumer group.

A full walk through of Redis, how a single threaded server is this fast, how it deletes expired keys in the background, how persistence and Lua scripts work, and how to decide between Sentinel and Cluster.

Every senior engineer I respect ships less code than the juniors around them. Two principles explain why, YAGNI and KISS. Here is what they actually mean, when you are violating them, and how to apply them.

We all know the ELK stack. But how does a log line printed inside a Kubernetes pod actually travel all the way to Elasticsearch? Here is the full journey, from stdout to docker logs to fluentd to Elasticsearch.

A harmless race condition met a well meaning retry policy and turned into an 8 second latency spike. Here is how I traced it, and the one line fix.

A practical guide to the DevOps concepts every software engineer—yes, even frontend engineers—needs to understand.

How mishandling Node.js streams can cause memory buffer leaks, stale data exposure, and how to fix them.

A breakdown of how I migrated from Heroku to a more cost-effective stack using Supabase, Railway while maintaining performance for calcont.in.

A reflection on my 1.5-year journey at Media.net—challenges, growth, and unforgettable experiences.