ISSUE 353 · September 14, 2021
Trends
The ‘Great Attrition’ or the ‘Great Attraction’?
A record number of employees are quitting their jobs and that's especially true in tech. Whether you're looking for work or looking for workers, understanding what's behind all the movement will help you make smart decisions.
McKinsey & Company
Sponsored Link
Leverage consumer insights data
Join AWS Data Exchange and thought leaders from companies such as Acxiom and BlastPoint. They will share how organizations from retailers to auto brands are using consumer insights data to reach new customers and drive real business change.
Tutorials, Projects & Opinions
Why data scientists shouldn’t need to know Kubernetes
While it may be useful to have full-stack data scientists, better tooling would help ensure that their time is well spent. In this post, Chip Huyen explores the evolving expectations of data science roles, current tools that help abstract away lower level tasks, and what's still missing.
Chip Huyen
Possession Is The Puzzle Of ⚽ Analytics
Great post about soccer's next big metric. Ultimately, "possession value" seeks to model the state of the game at any time and predict what that means about future rewards. For a lot of reasons, it's a non-trivial problem and there are a lot of models to choose from. Here's why.
FiveThirtyEight | John Muller
Gaussian processes From scratch
This first in a three part series introduces Gaussian processes and implements them from scratch using Python and NumPy. This is easy to follow with lots of visuals and code snippets along the way. For parts two and three, follow the links at the top of the post.
Peter Roelants
Intro to Bayesian Methods for Media Mix Modeling
Media mix modeling is a type of analysis that helps optimize how marketing dollars are spent across multiple channels. Linear regression is commonly used but here, Michael Kaminsky introduces an approach using Markov chain Monte Carlo (MCMC), which can be more powerful and flexible than linear regression.
Recast | Michael Kaminsky
What’s An OLAP Cube?
Great introduction to OLAP cubes. Slice, dice, pivot, and hypercubes — this covers it all. And along the way you'll learn how this pattern fits with a modern data warehouse.
Claire Carroll
TransformX Conference: Driving AI from Experimentation to Reality
Join Scale AI for our two-day, virtual conference featuring 100+ speakers and 60+ sessions. We’re bringing together a community of leaders, visionaries, practitioners, and researchers across industries as we explore the shift from research to reality within AI and Machine Learning. Register now to secure your free ticket.
// sponsored
Code & Tools
River
River is a Python package for machine learning on streaming data. Models process one observation at a time and can be updated on the fly. This is a general purpose library that handles regression, classification, and unsupervised learning.
GitHub | OnlineML
 |