Last year, while scaling an API service, I found myself reaching for Redis and EC2 auto-scaling groups far too early. It seemed like the “right” thing to do—after all, it was what I’d seen in tutorials, open-source projects, and most blogs. As a junior developer, I thought that using “production-grade” solutions like these was the most logical path.
At the time, it felt like progress. I was learning AWS, configuring auto-scaling groups, and managing Redis instances. But what I didn’t realize was that all of this came with significant overhead—both in terms of time and complexity.