ISSUE 409  ·   October 18, 2022

Insight

Internal tools make lousy startups 

Starting a business around the internal tools you already know and use may sound like a good idea but, for a lot of reasons, it's probably not. This is a great post exploring all the reasons why, including ideas for a better way to approach a data startup.
Benn Stancil

The State of Open Data 2022

The State of Open Data 2022 report summarizes the survey results from over 5400 researchers regarding their attitudes towards and experiences with open data. There's a lot here, including articles from data experts at prominent organizations around the world.
Digital Science

Sponsored Link

IMPACT 2022 | The Data Observability Summit

Join us on October 25-26th for IMPACT 2022, Monte Carlo's Annual Data Observability Summit. RSVP to hear from some of the industry’s most prominent voices, as well as the broader community of data leaders and architects paving the way forward for reliable data.

Tutorials, Projects & Opinions

Experimentation Platform in a Day

Great post that walks through the essential building blocks of an experimentation platform. In most cases, an experimentation platform is easy to start building and doesn't need to be complicated.
Deliberate Data Science | Dan Frank

How the Guardian uses NLP for quote extraction 

A lot of media companies are experimenting with NLP to "modularize" content as reusable elements for different formats, such as podcasts and infographics. This post explores the challenge and how the data science team at the Guardian approached it using spaCy, Prodigy, and a human in the loop.
Explosion AI | Ryan Wesslen, Victoria Slocum & Chung-fan Tsai

Quarto Questions

Quarto is an open-source publishing system that lets you create dynamic technical content using Python, R, Julia, and Observable. It's becoming popular fast, but if you're not already using it, this post is a nice introduction to Quarto that will help you decide if it's worth diving into and answer key questions along the way.
Occasional Divergences | Marc Dotson

The Data Liberation Project

The Data Liberation Project is a new initiative to identify, obtain, reformat, clean, document, publish, and disseminate government datasets of public interest. This is a project from Jeremy Singer-Vine, author of the Data Is Plural newsletter. It looks promising. 
Data Liberation Project 

How to make the most of item recommendations

Interested in learning more about item recommendations? Discover how Data Scientist Narendra Parigi experimented with complementary item recommendations on Adevinta’s e-commerce marketplaces. Read how to implement methodology, label data, create a pipeline and train the model. Read the blog
// sponsored

Code & Tools

Accelerating JupyterLab

The next major release of JupyterLab will be significantly faster than previous versions. Here's a look at what to expect.
Jupyter Blog | Frédéric Collonval

Career

New Opportunities

The Data Elixir Job Board currently has 35+ listings for remote positions, including roles for data scientists, data analysts, researchers, data architects, machine learning engineers, and more. The roles cover a variety of job levels, from mid-level to Director.
Data Elixir Talent Collective

Data Visualization

Mapping and Data Visualization with Python

This is an intermediate-level course for creating static and dynamic visualizations with spatial data using Python. It's free, self-paced and organized as a collection of Jupyter notebooks. It covers a lot of ground, starting with an overview of the Python visualization landscape.
Spatial Thoughts

How did you like this issue of Data Elixir?

👎    1     2     3     4     5   👍

Was this email forwarded to you?  Sign up here >>