Machine Learning legend Vin Vashishta swings by office hour to chat! Tonnes of awesome insight into what the future of data science is going to look like, why feature engineering can be dangerous, why a model is a hypothesis, and more!
Huge thanks to our sponsor for this episode - Cloud Academy! Go to https://cloudacademy.com/ and use the coupon code ARTIST for 50% off the monthly subscription fee for life!
[00:03:54] What I do as a mentor at DSDJ
[00:10:56] Vin’s perspective on the data science job market due to COVID
[00:14:09] Is data science going out of fashion?
[00:16:55] The two types of data scientists out there, one of them won’t survive
[00:21:50] You know, it's funny. It's got to be monitoring and production notes and stuff.
[00:23:25] Question on a project that an attendee was working on – clustering and topic modeling
[00:27:27] Saving models (to serve later)
[00:32:02] Is analytics data science?
[00:35:13] A philosophy less on feature engineering.
[00:40:45] Old school data mining and feature engineering
[00:46:02] You must validate your model
[00:55:15] Why is a model a hypothesis
[00:59:08] The importance of experimenting
[01:04:37] Is it ever OK to build a biased model?
[01:08:59] Preventing bad biases
[01:10:54] A philosophy of modeling
[01:15:24] Do we rule out deep learning?
Check it out and don't forget to register for future office hours: http://bit.ly/adsoh
If you want to interact with me multiple times a week, join Data Science Dream Job for 70% off: http://dsdj.co/artists70
Watch the episode on YouTube here: https://www.youtube.com/playlist?list=PLx-pFw_ty92wJoWzoO7WlfaM7iYB8_qjm
We were voted one of the top ten data science podcasts by FeedSpot - check it out here: https://blog.feedspot.com/data_science_podcasts/
I was on the Human Prosperity podcast, check it out here: https://www.youtube.com/watch?v=RbD3aV2TfKk