⚾️ The MLOps Big Leagues ⚾️

Before this meetup, I did some heavy binging on our guest Joe's youtube channel and I highly recommend you do the same if you are into this kind of podcast-style conversation around data science, data engineering and ML.

Joe had so many insights for us, from setting up the correct data foundation to start leveraging ML all the way to what his view of the current landscape looks like. Some of the gems from this talk revolved around choosing the right tooling in an age of uncertainty, how to be agile with your data and prepare for the worst, and the importance of the organizational piece in MLOps.

Check out the full conversation on video here and podcast here.

Nearing the End: All good things must come to an end some time. But for now that hasn't happened so let's enjoy these last moments and episodes while they last! Fabiana and the YData team have been so awesome throughout this whole series I tip my hat to them and give thanks for the effort they have put into making this series truly informative and entertaining!

For this episode we talk encryption. Machine Learning models are powerful but having them shared or not well secured, can incur massive privacy issues. So what can we do to make sure they are secure? What techniques are available? Ultimately what is encrypted Machine Learning and what benefits does it deliver.

To close 2020, we've invite Théo Ryffel, founding member and Crypto team lead of the Open-Mined community to explain to us the basics of Encrypted Machine Learning. It's a fascinating talk that will leave us wanting to explore more around the topic!

The Lowdown: I am not sure how it happened or how I found him or how I even convinced him but we had none other than the head of SRE for ML infrastructure at Google come on to the coffee session and talk all things tech, managerial, and ethics around ML these days. (I couldn't help asking him about the infamous email). Oh yeah, and I know there are a few out there that will be happy to know we got Vishnu to join us again. (Yes, he already has a fanclub)

Favorite quote: "Bad things happen to good data"

Couldn't have been more right about that one. Two main points he made that kept me thinking and then stimulated more questions were: the idea of ML "trustworthiness" and those of us interested in MLOps right now are either going to be the ones who write the books on it or are the first ones to read em!

Have a listen to the podcast here or check out the video here.

Also as a side note, Vishnu, David and I just recorded a coffee session with one of the authors of Machine Learning Design Patterns Sara Robinson, so keep your eyes peeled for that episode to drop in the week between xmas and new years. Oh yeah, and we will be giving away 3 copies of her book over the next couple weeks to those who give us their best horror story of a design pattern gone wrong!

Enjoy the break everyone! Hope you have a wonderful time with your family and loved ones, we will be back with more meetups January 6th 2021!

Ok OK Ok, You didn't think I would let you go that easily, did you? If you are jonesing for more MLOps I got one last thing for you to finish off the year. I put together a holiday animation video that explains what a feature store is in a very Christmas kind of way. Check out the 3-minute clip of what I felt like a good metaphor for a Feature Stores could be.

If you have any cool metaphors for what ML Monitoring would look like, shoot me a line cause that's next up on the list of animated metaphor videos.