SF Data Weekly - Simple Models, Design for Neurdivergence and Actual ML Work

How do you get better as a data scientist? Do you learn math to build intricate models? No! Analyze the data first to see if you can use simpler models.

One way to expose Redshift data to partners is via a REST API. This example shows how to use Lambda and SQS to expose a REST API to partner pipelines without exposing the direct Redshift connection.

If you have outgrown Salesforce Data Loader and the Salesforce Data Import Wizard - you need a more scalable solution. Check out Integrate.io’s Salesforce capabilities. [Sponsored]

AQUA is the Advanced QUery Accelerator for Redshift. This piece shows a test of a RA3.xplus cluster to show the impact of AQUA.

A good step-by-step showing how to analyze data and create a model to predict house prices.

Data and Reality by Bill Kent is a book on the philosophy of information and how it should be represented. This piece explains why it’s worth reading.

A real data scientist shares his work from last year. Spoiler alert: some of it wasn’t data science.

“Neurodiverse” covers a number of neurological conditions including ADHD, Asperger’s and Autism. This piece concentrates on the visual factors important in conveying information to the neurdiverse.

This piece picks a number of articles related to data visualization

The bigger picture of data visualization: how to create a narrative with your visualizations.

How GE took their engine wash analytics process from 2 day runtimes to a little over two hours.

A four-minute video summarizing the BI developer role. Here are our jobs for the week: