| I think the core problem is that there isn't a reliable way to assess a person's SWE skills. Two people can look the same on paper but have wildly different capabilities for solving problems with software. So many companies try so many different ways to establish competence during application and interview but they are all badly flawed in some way. Add to that the fact that no two SWE teams are the same, and to be effective as an IC you need to be effective as a team member, and it gets really hard. The true test really comes during the probation period (which I think of as the real interview) and if all things were possible, I'd hire many more people off the back of resume and interview with intent to retain those that are able to perform at the level they pitched themselves. There's always a bit of "leveling up" as people climb the ladder, so a little bit of over-stating one's own skills is to be expected, and that's where the core capability of "being able to learn quickly" gets proven. Unfortunately budgets and my own capacity don't allow me to take this approach. I'd also say that as someone who has led or managed SWE teams for a long time now, SWEs do love to argue for their favourite bit of tech. After all, if they've used it before, they'll be more productive with it (perhaps). However that drops us right back to "the list of technologies you've worked with" mattering more than "your ability to pick up skills". I do wish there was a better quicker way, that could not be games or bluffed, to hire people. Many startups are trying to solve it, but none (AFAIK) have scratches the surface. I think (IME) that getting a job is complicated because (a) it's very hard to determine a SWE skills by resume and interview (b) lots of people over represent their skills (c) hiring is probably the most critical thing an organisation does (d) SWE has become more complicated over time, rather than less, leading to proliferation and speed of introduction of new languages, frameworks, tools and libraries. |