Summary of The Data Janitor Returns | Daniel Molnar

00:00:00 - 00:20:00

The Data Janitor Returns is a video that discusses the importance of data janitors in the advertising industry. It suggests that if someone is seeing too much data janitor activity, it may be time to move on from their advertising career.

• 00:00:00 The speaker at this conference talks about how data janitors should always do their homework and double-check their work for accuracy. He also stressed the importance of having clear, programmatic KPI definitions.
• 00:05:00 The Data Janitor video discusses how to identify and fix issues with data in a company, including identifying when data is wrong and how to do a B test to see if a change in strategy is warranted.
• 00:10:00 The speaker discusses the need for data scientists to have more engineering skills in order to build foundations for machine learning hierarchies. He also references a blog post by Pete Worden, a well-known leader in the field of machine learning, in which he discusses the issue of compression being a good use case for deep learning.
• 00:15:00 In this video, mentor Daniel Molnar discusses how distributed computing is not a solved problem and how data engine is someone who embraces his dirty reality. He also advises that individuals try to get cloud-agnostic, as that is the "rage" among businesses. Lastly, he mentions that while Ascher-cut salaries are going down, cloud computing is still a viable option for those looking to save money.
• 00:20:00 The video discusses the effects of data janitors on the advertising industry, and how to identify and deal with them. It suggests that, if someone is seeing too much data janitor activity, it may be time to move on from their advertising career.