|
KOTEX GOD OF BLOOD posted:I've been learning some crude data science / data analysis stuff in my job, but I think it would be really helpful for my career to take it more seriously and git gud. At this point I'm decent with Excel, know a schmear of R, use ChatGPT to write Python I can use, and am getting familiar with Postgres. There are a litany of online courses out there, but I'm curious if people have had better experiences with one or another. Work will pay for this so money is no object. Hey so this is a great question. The quick answer to this is build as much stuff as you can. Answering questions from data, building dashboards, running experiments, building models are all good things to look into. Both these paths are pretty self-starting, so having experience doing as much as possible will serve you well in your career and your day-to-day duties. If work is paying, you may as well get the best certs you can from Microsoft, Amazon, Google, or a degree from a good university if you have time. In more detail, both data scientists (DS) and data analysts (DA) work with data but the skills they use and the goals they work towards tends to be different. This varies considerably across orgs and industries so YMMV. Some orgs use the titles interchangeably, if that's the case then usually everyone is an analyst. DAs tend to be "business-focused". They answer questions for stakeholders, whether ad-hoc or regularly like building a dashboard. The tools they use tend to be most impacted by how complex the business is. Less complex sticks with spreadsheets, maybe a graphic or two. More complex means you're querying databases or data warehouses, building complicated dashboards with Tableau, even doing some data modeling (organizing data stores). Some usual pitfalls of being a DA are sometimes having less autonomy, and that you're an in-between to business and tech - companies have a tough time dealing with this sort of organizational stuff a lot. DSs tend to be more focused on bringing something to operation. Think designing an experiment, answering a complicated statistical question, or building a model to predict something on-the-fly. DS also sometimes include methods of data collection. Most notably is it seems like DS has the highest expectation of autonomy. They write code, queries, and systems focusing on things happening automatically. The downside of being a true data scientist is its hard dealing with everyone's expectations that data science is magic or fluff (there's no in-between) Phraggah fucked around with this message at 00:45 on Nov 24, 2023 |
#
¿
Nov 23, 2023 11:10
|
|
|
|
| # ¿ May 21, 2024 21:19 |
|