Has software engineering shifted from “building to last” to “building to replace”? Earlier, we'd build a system, take it to production, and it would run for years with small enhancements and maintenance. Now. every few years there's a push to rewrite everything with a new tech stack, often because the existing system is considered "outdated" or "not sustainable." Are frequent rewrites driven by real business needs, or are we too quick to replace systems instead of evolving them?

My manager wants me to mentor our new hire next month. I’m a little stressed about it because I still feel like I’m still just winging it most days, even after a year. Is it normal to feel underqualified when you first start mentoring someone?

Final-year Software Engineering student based in Perth, WA, actively applying for internships. Would love any advice from engineers on what companies look for in junior/intern candidates — especially for AI or software dev roles. Also curious what the Perth tech scene is like for new grads!

What’s your take on oversized Pull Requests? Knowing that it’s the norm of AI assisted development, automation hasn’t still caught up with humans reviewing the changes.. I would and still reject overly large PRs for reasons outside of testing and fixtures.

I keep blowing past my own estimates, and I'm trying to figure out if there's a way to get more accurate. Every time I think I've scoped something out, the actual work takes way longer than I planned. What's helped you close the gap between what you think something will take and how long it really ends up being?