All Hands on Data #45
All Hands on Data #45 introduces another new sailor to the crew, Angel! Check out his first submission below!
A Data Scientist Breaks Down All 10 Taylor Swift Albums
For all the Swifties out there with a data science inclination, here’s the article for you. The author touches on several interesting aspects and presents great imagery to help visualize the changes the artist has undergone through her music. This piece knew it was trouble when it walked in… - John Forstmeier
Polars vs Spark. Real Talk.
I found this interesting because I agree that Polars is one of the more exciting developments in the data space because it addresses a true pain point for pandas users. However, the existing tooling around big data in particular (Spark in this case) is not going anywhere any time soon. Polars is great for a large data set on your local machine, but if you need to process 10’s or 100’s of TB’s, the best bet is still Spark. - Wes Poulsen
Zero-ETL, ChatGPT, And The Future of Data Engineering
Where is data engineering headed next? Moses talks about the front-runners, who they are, what they are, and their pros and cons (in a writing-style that is as engaging as the material). - Katt Baum
What Does Democratizing Data Mean? Unlocking the Power of Data Cultures
The author brings up good points for why we should make data accessible, understandable and actionable for a wider audience. The article explores strategies for successful data democratization, tools, and its future prospects while acknowledging challenges such as data quality, governance, privacy, and culture. - Jon Davidson
Creating Healthy AI Utility Function: Importance of Diversity - Part I
As AI becomes increasingly more prevalent in our day-to-day world, many stakeholder are giving their input on how AI can help their business perform better. With this, brings the concerns of ethics into play when it comes to AI. This article dives into the importance of diverse teams to lend multiple perspectives on AI’s uses, as well as the importance of minimizing biases by carefully considering data and algorithms used to train AI. - *Reed Cowan *
Data Council 2023 Highlights
Blake and I went to the Data Council event in Austin, TX last week. We broke down what we learned from our favorite sessions. Check it out if you missed the conference! - Angel Catalan
LGBTQ+ bias in GPT-3
The article “LGBTQ+ Bias in GPT-3” explores how GPT-3, an AI language model developed by OpenAI, may produce biased or offensive responses due to its training on internet data that includes discriminatory content. The author provides examples of biased outputs related to sexual orientation and gender identity, discusses potential solutions like fine-tuning and human review, and emphasizes the need for transparency and ethical AI development to ensure inclusivity. - Steven Johnson