8
votes
What did you do this week (and weekend)?
As part of a weekly series, these topics are a place for users to casually discuss the things they did — or didn't do — during their week. Did you accomplish any goals? Suffer a failure? Do nothing at all? Tell us about it!
Fairly chill week (and weekend prior to said week). July 4th weekend was last week and my city had a big fireworks display that ended up being quite disappointing unfortunately. Wasted 3 hours of my evening trying to both get there and then ended up seeing just flashes of light haha. That weekend wasn't a total wash though, went up north from my city and explored a famous lighthouse which was quite fun.
Last month I was pulled into this new AI org at my company and they had their on-site (corporate lingo for gathering) for this new org this week. I have been a handful of on-sites in my short career so far and I have never seen one where people were less eager to be there than this one. This on-site was also a waste of time, there were multiple presentations that just regurgitated info we'd gotten a month prior. I'm actively trying to switch teams but the ball is in the court of the hiring managers so I can't really do much.
Sorry for my usual work rants in this weekly thread, it's become a bit of a pattern haha.
Just out of curiosity, why are you trying to switch away from AI? I work in the field as well and even though there's a lot of hype there's some really cool stuff going on and we're booking some (small) successes!
I have nothing against the AI industry but the AI work I'm doing is none of the cool, fun parts imo. The AI work I'm doing is just generating data that would then potentially be used to train models. It's pretty menial work without a real feedback loop.
⚠️Rant incoming, sorry ⚠️
In standard software engineering, I start with a ticket where XYZ metrics are low. I work on things that would hopefully improve those metrics, deploy those changes, and then watch how my changes affect XYZ. If it improves, my job is done and I move onto the next thing. That feedback loop is critical and I really thrived on it. With this AI data generation work, all I do is throw my code into the void. The only signal I get is if my code is high-quality enough to be used. I am not told what makes my code high-quality enough to be used. The only thing I am told is that I have to generate two chunks of data a week. There is no feedback loop. I don't even know how career progression is supposed to work given everyone, from people with 6 months experience to 16 years, is doing the exact same work.
The new AI org I'm in is also just impressively unstable. Major things change every single day. Org updates shared in the morning are outdated by the afternoon. There are different "tracks" of work you could be assigned so that a large breadth of data is generated and you're supposed to be assigned a track for the week on Monday. Often times, they (leadership) will just change your assignment mid-week and go "oops, now you have to work on this". For a company that prides itself on corporate transparency (which still exists in other orgs), this new org is frustratingly opaque and leadership doesn't seem to really understand why people are upset.
I'm glad I still have a job and still earn what I do even with my reduced workload but it's just not what I want to do.
Thanks for sharing that does sound like a very frustrating environment to work in! What's the goal of the code that you're submitting? Just to train an (internal) model?
It's a shame that they don't give you more autonomy, AI is such a good tool to allow a motivated individual to have an large impact on the organisation.
In a way it's also a perfect Dilbert-esq scenario. You're working on something incredible high-tech in a management structure from 80/90-ties corporate America (low autonomy, continuously changing goals, "corporate transparancy", management not understanding feedback, etc).
Yep, we're training an internal coding model, part of the work is seeing how well our internal model does on certain tasks and comparing it to the existing industry models. Definitely agree that AI can give you more autonomy. I was a pretty well-oiled machine back in March/April when I had a few workstreams going on in parallel and AI was helping me stay on top of all of them.
It really does remind me of Dilbert haha. Funnily enough, leadership make an apology a few weeks ago about how they handled this re-org and vowed to change a few things. The first thing they called out was the snacks in the office? I can't recall a single person really complaining about office snacks so I don't know what possessed leadership to address that first lol.
Had a hypertensive episode this week and ended up in the ER. I was in that "too sick to go home" but not high priority category. I tried urgent care first but my BP was dangerously high, just not high enough to avoid 4 hours in a mostly empty waiting room.
It kind of came out of nowhere...ish. I'd talked to my doctor and m had some slight highs before and we were just monitoring but this was me feeling my teeth as my whole side of my head was pulsing. I think I maxed at 212/106. Do not recommend. I've never felt anything like it.
So now I am trying to feel fully recovered while not straining myself while still doing things. I follow up with my doctor tomorrow and now get an ER bill on top of everything.
But today I sprayed weed killer on everything that has taken over my yard. So maybe the weeds turning into trees won't tear my fences down and my driveway won't become grass. It's not my preference but I can't afford to pay someone to do it manually.