Rust was once a language just a few enthusiasts mentioned on forums a few years ago. Having a language that was genuinely fast, memory-safe without using a garbage collector, created a fair amount of buzz, but a language like Rust was still too niche for most programmers to use. That is beginning to change. Searching through some of the latest job postings for the backend or systems positions shows a significant shift toward Rust, as it is becoming more heavily demanded compared to just a few years ago.
Rust is now delivering on what it set out to do. As with languages C and C++, a garbage collector is not required when achieving memory safety. Unlike the garbage collected languages that incur a huge performance overhead, Rust achieves this. For gigantic applications that are supposed to be used at scale, such as infrastructure and backend systems that are performance critical, Rust is truly unique in being able to achieve this. This is really the primary reason a number of organizations have begun implementing Rust compared to other languages.
The main obstacle in doing so is that Rust does have one of the most challenging learning curves. New users often do not grasp a number of concepts in Rust annoyingly quickly, and seasoned developers coming from other languages often find errors during the initial weeks of development, despite assuming the errors to be completely incorrect.
For teams having to consider performance, time spent learning could be a legitimate reason to avoid using Rust in development, but otherwise, its appeal should only keep increasing.
What has changed for a lot of companies is not the lowering of the learning curve. What has become harder to ignore is the cost of not using Rust for certain critical areas. This is especially true with the recent increases in security and reliability concerns throughout the industry. This is especially true when considering if it is worthwhile to learn Rust. The answer is it depends on what you are building. For general web development, it is still overkill. However, for any situation where a memory bug in production is extremely costly to get wrong, it is becoming the clear choice rather than the uncommon choice.