— Insight —
Nice little essay about the life of a data geek in a pandemic. "Every day begins in the same way. I get up. I make my coffee. I look at the data. Everything about this is absurd..."
Once upon a time, there was a man who thought love was a maths problem...
— Tools and Techniques —
Five years after it was published, a well-known Google paper is still cited as a resource for understanding technical debt in ML systems. In this post, Matthew McAteer revisits the original paper with a modern perspective and offers a new set of best practices along the way.
Great post by Sean J. Taylor that explores five key properties of metrics you should consider when developing a measurement strategy.
This new open-source project from Microsoft uses differential privacy to add noise to data, which helps to prevent the disclosure of sensitive information. This introduction to the project shows how to get started with simple use cases.
Good decisions require good data and when you're working at Uber's scale, ensuring quality data is nontrivial. This post explores Uber's Anomaly Detection System and how a new solution uses statistical modeling to automatically monitor and flag anomalies at scale.
TLDR is a daily newsletter with links and TLDRs of the most interesting stories in tech, science, and programming!
— Data Viz —
Color plays a key role in the analysis and communication of data, especially in the sciences. Here's a good overview of how humans perceive color, problems with common color schemes, and how new tools are helping to make sure that color is used effectively.
Nice collection of tips and resources for developing data visualization skills. Covers things like how to choose projects to learn from, specific data skills to develop, design and storytelling.