Should I do a doctorate straight after my master or work for a while first?

Work first. There're many reasons for this:

1. It's not true that you can't work with cutting-edge technologies without a PhD. You certainly can, and it might even be a better use of your time since you avoid the administrative processes that you must do with a PhD. Check out the various graduate programs that employers offer. There might be high entry requirements, but if you can do PhD studies you can clear that bar also.
2. You likely earn more. It's not just for the next 3-5 years that you need for a PhD. If you do a PhD, after you graduate, you'll need to find a job. If you get a job now, you'll already have a job. If you do well you might even command a higher salary in 5 years' time compared to entering the market as a fresh PhD.
3. You might find you don't need a PhD. This could especially be the case if you work with other PhD-holders. You might find that you have the same duties, or that you're already capable of doing what they do. In this case a PhD is not very useful for your career and you might as well stay put.
4. If #3 doesn't happen, then you have a better idea why you're doing a PhD. You know what you want to learn, why you want to learn it, and how that skill is going to be useful for you after you graduate.

It's my observation that people who work first and then go to a PhD are much more likely to have thought seriously about why they're studying. That is a good thing. I'm not saying you shouldn't do a PhD, but you should have good, clear reasons for why you're doing it to avoid possible future regret.

On a cautionary note regarding working first - I decided to work first, as there are a number of advantages. I am now 33 and have finally applied a second time this past month (the first time was 8 years ago, and much less coordinated). I no longer have strong academic connections which I can use as references. Ultimately, life got complicated and busy, and it took me 12 years to apply for the doctorate I've long sought.

I believe I am no longer an ideal candidate as a result - while I've extensive work experience, am (more) mature and know how to work hard, I have been too busy working to demonstrate serious research ability. Additionally, should a program accept me, I have fewer work years left to return to the field than a younger individual with stronger references.

As a further note, I frankly don't need a PhD - only if I'd like to continue along my current career track (web developer, just switched to embedded systems due to tremendous preparation and good fortune). My father (an academic himself) calls it "an expensive hobby", and additionally points out there are a lot more PhDs than jobs for them.

I still feel there are substantial advantages to working for a time first - just, be aware that you will need a specific actionable plan for keeping strong ties to academia, or you will face increased challenges getting accepted in the future.