Hacker News Reader

Comment by Waterluvian | Hacker News Reader

My #1 issue as an engineer is running out of steam. By the end of a feature development I really struggle to force myself to end it properly. To not rush the testing. To update all the documentation. To fix the discovered quirks “properly.”

I think it may be related to ADHD because I’m getting paid the same and there’s no rush to move onto something else. The work environment is quite good.

My current coping mechanism is to work on two (but never more) projects concurrently and swap between them every few days.

View on HN

I have found success in using your "coping mechanism". I usually alternate between 2 projects that need my "engineering cycles" and then have N that need my "management cycles", which I don't find as exhausting and don't drain "steam" as quickly.

In bodybuilding, this is called "periodization", I'd say that's a pretty accurate term to describe this.

this sounds interesting to try for me. Software engineering alone wears off for me after a while on a project to the point I simply cannot type a single line for it if forced. Managament is where I am really good at coming up with ideas/plans/strategies but the day-to-day following-up (including staying calm and friendly in the face of very slow employees sometimes) also is soul crushing and depressing after a while. My sweet spot somehow is in the middle an hopping between projects, mentoring engineers and also doing my own projects on the side or even spearheading initiatives... sometimes.

What cycle times/scheduling works best for you? how do you deal with the dread of picking up a topic again after letting it go mentally?

To be honest I go by feel. I usually can't spend more than 3 full focus days on something without getting burns.

No ADHD here but all of that is perfectly relatable.

The last 10% of polishing is hard, and the 10% after that is even harder.

Working on multiple projects helps me too.

Have ADHD and this matches up perfectly with my experience. Finishing feature work off (especially things with a long timeline) is difficult and I tend to also cope that way, working on multiple things at once because it’s difficult to tunnel on any one thing.

> to end it properly. To not rush the testing. To update all the documentation

A trick I've learned is, when possible, to start with tests and documentation.

If you can, implement code reviews in your integration pipeline. Someone pointing out my lazy shortcuts keeps me honest.

Reply on news.ycombinator.com