More and more lulls with more and more years of experience. I hit the gym more, socialize more, cook more extravagantly, take walks more often etc. The most important thing was to train myself to not give a damn when people were making stupid decisions at work that were going to bite them N months down the line during those lulls.

This kind of implies that you're crunching and then 'recovering'. That may or may not be something you have any control over - there's a lot that goes into creating an unsustainable 'sprint', and probably 99.8% of it is not related to actual developers or code - but ideally you would be using these 'lulls' to try to pull stuff out of the next crunch so maybe it won't hurt so bad.

In reality, if I'm coming off of a bad crunch, I do anything I can do to avoid burnout. Sometimes that's 'fun' backlog items or research for future features or something else I'm excited about, sometimes it's studying for certs, sometimes it's cutting slack (@cianuro@programming.dev watching Netflix feels familiar!). But again - whatever it takes to recharge my batteries and feel less bitter and shitty.

The most 'sure' sign that I'm coming off a crunch, though, is that I start reinforcing work/life boundaries. "It's 5p and I'm logging off and I'm not going to think about work shit willingly until tomorrow."

I’m at a startup so I rarely have lulls. In previous experiences we even watched Netflix in front of our manager lol. I also used that time to study to get a better job 😉

Lulls as in nothing to do? No, I don't have those.

I normally try and do "fun" work. This largely depends on how autonomous your job is. I was a PhD student doing research for a company and I received very little oversight for 3 years.

The supervision I did receive was great though. They understood needing to take a break and slow down. At those point I would generally read papers, watch PyData talks (highly recommend them, like inspirational ted talks for data people), or contribute to open source to learn about new tools or design paradigms.