Front-end lead here, building large desktop apps for my company's internal use. I've never had a web-app that failed due to performance. I can't recall any tha

Front-end lead here, building large desktop apps for my company's internal use. I've never had a web-app that failed due to performance. I can't recall any that ever have, except Facebooks' attempt to use HTML on mobile. UI performance has little to do with algorithms, and only occasionally requires opening the profiler to hit 60fps with some tweaks to bone-headed nested loops. Or maybe reconsider your use of Angular! Front-end apps fail because long-lived mutable state is a powerful engine of complexity, and algorithms do not solve complexity. I am constantly telling my team (CS back-enders prepare yourself!) to ignore performance concerns. O-whatever has zero-impact on delivering simple, extensible, bug-free web apps that please its users. But admittedly, crafting good software isn't as objectively clear as implementing a mathematical function that makes your server code perform. The proliferation of front-end frameworks and compile-to-js languages indicates that writing FE software that works remains a far greater challenge than writing FE software that hums. To that point, maintaining a popular github repo for a library or app that demonstrates your ability to write, maintain, and extend FE software is still the best sign that you know what you're doing, and that you like what you're doing. Conversely, knowing a lot of algorithms as a FE candidate tells me nothing about whether your first PR will be your last. On a happy note, Netflix didn't ask me a single question regarding algorithms! Their interview's focus on OOP made sense, even if it didn't fit with my own focus on ClojureScript and FP.