I'm in my late thirties and starting my first 'proper' engineering job next week. The career change process took me about four years all told, but life gets in

I'm in my late thirties and starting my first 'proper' engineering job next week. The career change process took me about four years all told, but life gets in the way and naturally YMMV. I was working in administrative and funding roles for non-profits and hated it. I'd been doing it since school and had no idea what else I could do. I ended up going back to study, choosing a social science Masters with a strong quantitative focus. I really clicked with the code side of things. I'd done bits of BASIC, VBA, and Pascal as a kid, but nothing more than ridiculously over-worked Excel books since then. I actually picked up Python at school before taking a module in it. A friend who knew I was interested in learning to code needed some data from a website and asked if I'd be able to scrape it. I said "No, but I think I can work it out." I still recommend a scraping project to anyone trying to learn to code. You need to have this very basic understanding of http requests, HTML, and flow control in whatever language you're using. And you get something out of it at the end. I love reading programming books, but I find them best for consolidating and pushing forward skills. I feel I've learned best when working on projects, and personally didn't get on with things like Learn Python the Hard Way or online courses. May just be my learning style though. I've built my skills though finding lots of little projects to do; either as part of my course, or later for a (sort of) related job, or in my free time. Whilst reading is definitely important, try and find stuff that interests and excites you that you can find an excuse to code with. Things I've really learned from include: - Building a tool to create corporate networks from the UK Companies House API; - Scraping and analyzing sales on a darknet drug site; - Getting R code running in Python (Python sucks for survey analysis, R sucks for day to day use); - Reverse engineering a police incident map to get at the underlying data; - Working out the distance, by road, between any given point in my city and a fire station; - Using the Blender Python API to make custom 3D-printable bases for my wargaming minis. Good luck! It's daunting changing career and has been a slow and frustrating process for me sometimes. Totally worth it though. All the best to you :-)