Quarantine Day 38. Something I’ve started doing (again) these past couple weeks is establish a prioritized to-do list every morning. Ideally, I’d do this the night before (i.e. what I want to do the next day), but I’ve found that (a) I’m almost always too tired at the end of the night to plan, and (b) when I do make a to-do list for the next day, I almost always have eyes bigger than my mouth.
Apparently, I’m not the only person with daily to-do lists that’re just outright unrealistic. One rule of thumb I’ve seen time and again is that you should only have three items on the list. You only have three slots, so you’ve got to pick wisely. Then, by the end of the day, you have to have finished them.
Unfortunately, this hasn’t exactly worked out for me. One of my three items (during the workweek) is almost always “get work done.” Then I’d flip over to another daily to-do list, with its own Three Things To Get Done, that’s specifically for work. In some cases, some of these work items spawn even more sub-work items like an NPC requesting an endless chain of mandatory quests.
Of course, I already know the solution: choose even less than three things to do every day. Just kidding. I wish I could do that.
What would I do if I won the lottery? Where I won just enough to be completely financially independent, continue my current lifestyle, but not enough to throw cash at people from a moving car (or at strippers?).
This was the question that Khoa and Joe posed during lunch.
Actually, I wasn’t sure if I was even going to make it to work lunch (don’t worry, this is a completely casual hangout “meeting”). I’ve been literally pulling hair out trying to figure a bug the past couple days. That was how my day started: with lots of way-too-long hair on my table.
But sure enough, by the time lunch rolled around, it was confirmed that the bug wasn’t even mine. So much wasted time. But that’s how it goes.
As for what I’d do, it turns out that I wouldn’t really do anything different. I mean, wasted time on stupid bugs notwithstanding, I really do like my job (and I really do like my current company too!). And one thing that I like about my job is the fact that time off is flexible: whether or not I win the lottery, I’m still want to thru-hike the PCT, Continental Divide, and AT some day. I still want to do a cross-country bikepacking trip, either here in North America, Asia (as I mentioned in my New Years Resolution this year), or New Zealand (after I missed the Tour de Aotearoa this year), and the only things holding me back are my lack of cycling-specific fitness and even basic backpacking experience — not my job or my finances.
Apart from outdoor adventure-y stuff, I’d also consider going back to school for a PhD, though I’d choose my academic advisor verrrrry carefully. Or I might start my own consulting business, choosing clients and projects that I want (these would almost certainly be non-profit clients and projects).
Did I mention that this was Release Planning Week? When I first joined my team a little over a year ago, I was surprised by how unstructured planning was, at least on my team. Prior to Salesforce, I’ve only worked in academic settings and a startup. At the startup, we’d have sprint planning every two weeks: at the end of every two weeks, we’d create stories, point them, commit as a team to how much we could close by the end of the next sprint, then get to work. The whole shebang. The whole thing was pretty rigorous.
But now, even though two-week sprints exist, planning actually only happens every 3 months. It kind of makes sense. It also kinda doesn’t make sense. Because there’re so many moving parts and so many teams that depend on one another, it make sense to make planning somewhat longer term; there’s just that much more overhead required to align all the different teams. But, the trade-off is that there’s a lot less tolerance for unknowns built in (for example, the vast majority of bug-fixing work is unanticipated, and there’s almost always some unforeseen work that pops up when working on almost any work item.)
It’s even worse for data science: the very nature of the work involves uncertainty (is the model going to work? how many times will we need to rearchitect it to get it to work?).
All of this basically meant two things: (1) planning had to be very deliberate (we really had to try to anticipate what would come up over the next THREE months), and (2) execution that plan had to be rigorous.
I remember after my first couple releases, I told my manager that I was surprised how well everything worked despite planning not being as deliberate as I thought it should be, and execution not being as rigorous as I was used to. “Everything just works!” I said.
Everything still “works,” and I truly think that this is a testament to quality of my team. But, it’s also become clear that things aren’t working as well as they used to.
A huge part of this has been the increasing maturity of the product I work on. When I joined the team, the product was just exiting its experimental, proof-of-concept phase. My early contributions toward simply containerizing the product and increasing test coverage made immediate impact. Tweaks to the Tensorflow model often yielded exciting new findings that paved us toward clear decisions.
But as my product has trekked toward public availability, our needs have changed. I’ll probably expound on this in another entry (this entry is already growing too long!), but suffice to say that I’ve made moves to tighten up planning and execution.
This is something my manager started noticing a few months ago, even though I had really started from the very beginning. But because of various reasons, I didn’t really want to push for radical change — at least too quickly. Now, she explicitly called on me during a 1-1 to take more leadership; I had been working toward greater buy-in from a couple team members, and now she had proactively taken care of it.
What I’m saying is that this is the first release planning that I’m owning end-to-end. It’s taken up a bit of time over the past couple weeks, but I don’t regret it at all. In fact, I surprised myself by enjoying it. And best of all, I think my teammates actually enjoyed the process a little more this time as well.
Jackbox Party Pack 6. Now for something different. After I wrapped up the day’s work and meetings, I read a little more Murakami then readied myself to run.
Then I realized that, hey, I had a game night tonight — a follow-up to the game night on Wednesday. But, this time, we’d have 9 players instead of 4, and we’d stream on Discord instead of playing on Spendee.
If there’s one thing I like about self-quarantining, it’s the fact that I get to see a lot more people that I usually wouldn’t see that often (with the tradeoff that I don’t see people that I usually see often, as much.)
In this case, the new folks who I haven’t seen in at least a year included Di-Annie (Christie’s sister), Annie (whose camera was off), Jin, Michael (whose camera was also off), June, and Kevin (who I don’t think I had met before). I think my biggest surprise here was Di-Annie — she’s been working out!
It was a little weird playing Jackbox with people I don’t hang out with often (since a lot of the games involve understanding other peoples’ sense of humor), but I definitely had fun. The whole session lasted some 3 or so hours, followed by another hour of just banter, conversation, and a promise of having more game nights in the near future.
New Posts
- Myanmar After Eight Years (migrated from FB)
