Part of why I hate the term "insights" as a term for what data teams produce is that it's not the early 2010's anymore--no one cares about artisans producing bespoke, handcrafted things
Show this thread
-
Distributed computing and sprawling commercial codebases mean that working as a SWE is more than sitting down and plugging away in an IDE. Everything is connected and if you're careless about what you commit, your infrastructure itself can kick your ass
Show this thread
Data is exactly the same way. But unfortunately, using data in a commercial setting is a lot closer to that lone SWE in an IDE, except its a DS working in a Jupyter notebook or an analyst building a dashboard
-
-
DS and analyst work IS contributing a system. That system might be more abstract than an app or website, but the work of one DS or analyst are connected to each other and can absolutely have an impact on the work of their colleagues
Show this thread -
SWE is scaled by trying to automate infrastructure coordination, management, and risk. Infrastructure is to SWE as data is to DS. Not managing your data in an automated way invites coordination costs and risk, and it naturally leads you into the insights trap.
Show this thread -
Btw, if it seems like I'm talking about what makes SWE teams effective recently, it's because it's part of my job! If you'd like it to be your job too, I'm currently on the lookout for a tech lead :)https://smrtr.io/5BvSG
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.