Designing a Generation Pipeline — A 4-Axis Content Model, Queue-Based Scheduling, 3-Layer Attribution
Design decisions behind making content without a bot signature, publishing without SaaS, and measuring views → clicks → conversions

Search for a command to run...
Design decisions behind making content without a bot signature, publishing without SaaS, and measuring views → clicks → conversions

The data-acquisition layer of a marketing agent: mapping curl, headless Chrome, and userscripts onto an anti-bot threat model

"Make it faster" sounds like a vague mandate. It isn't. Performance work turns out to be one of the most learnable parts of backend engineering, because it almost always comes down to the same three m

42.8% HTTP error rate, 9 minutes into a load test. The culprit was a single updateMany line carelessly written during AI-assisted coding. 1. Discovery — "Why are we doing 9.5 TPS when the target is

A reproducible 7-VM clone of production, k6 + a Go WebSocket receiver, and the actual TPS our send API breaks at.

How I discovered a structural deadlock risk in KeyDB and migrated to Valkey Sentinel HA — without changing a single line of application code

How we rebuilt our meeting recording system with 5-minute savepoints, streaming uploads, and a fully automated STT-to-summary pipeline — solving browser memory limits, tab crashes, and Korean speech recognition along the way.
