The launch is the first thing a user experiences and the first thing they judge. What actually happens in the seconds between a tap and a usable screen, why almost all of it runs on one thread, and how to find the work that is making them wait.
Isolation levels are usually taught as an abstract table of names and guarantees nobody remembers. They are far clearer explained backwards — through the concrete bug each level allows, from dirty reads to the lost update that quietly corrupts your data.
Transaction Isolation Levels, Explained by What Breaks Read Post »
Indexes are the first thing people add for performance and the last thing they audit. Which index types actually earn their place, why the column order in a composite index decides everything, and how to find the indexes silently costing you on every write.
A naive request counter is easy to build and easy to bypass. The rate-limiting algorithms that actually work, why limiting by IP address quietly fails, and how to layer defenses so an attacker cannot simply route around the one control you built.
JSON Web Tokens are everywhere and misused almost everywhere. The predictable mistakes: storing them where XSS can steal them, assuming you can log someone out, algorithm confusion, and reaching for JWT when a session would have been simpler and safer.
Every React project reaches for a state library early, often before it needs one. The distinction that dissolves most of the problem: server state is not UI state, and most of what teams put in a global store is really a cache of the server.
Forms are the most common interactive element on the web and the most consistently broken. Validation that fires at the wrong moment, errors a screen reader never announces, autofill sabotaged by clever markup. How to build one that respects the person filling it in.
API versioning sounds like a URL-format debate, but the real problem is changing an API without breaking clients you do not control and cannot redeploy. What actually counts as a breaking change, and how to evolve an API for years without a v2.
There are only two hard things in computer science, the joke goes, and cache invalidation is one of them. The caching patterns worth knowing, the staleness each one accepts, and the stampede that takes down services when a hot key expires.