Is Gemini’s New Data Science Agent Useful? Here’s The Truth

Author(s): John Loewen, PhD

Originally published on Towards AI.

Testing Python code creation and distribution in Google Colab

In Google Colab, Gemini makes it possible to go from a plain-text instruction to a functional, multi-step notebook — without switching tools.

In other words, you can now prompt a Jupyter notebook to write itself.

This includes the full workflow of reading a dataset, cleaning it, filtering by year, and generating an interactive data visualization using Plotly (for example, a choropleth map).

And even better, you can copy the notebook from Google Colab straight into a Github repository.

Need some proof of this? Let’s work together using a dataset of world happiness scores to demonstrate the ease at which this can be done.

An uploaded dataset and a single prompt can generate everything: file upload logic, data inspection, filtering, and data visualization.

The entire process runs inline, in a single Colab notebook, with zero configuration and minimal manual coding.

You write the prompt, Gemini writes the notebook — here’s how it works.

To start off with, all you need is a Google Account and can use Google Colab for free.

Google Colab is Google’s cloud platform for writing and executing Python code. It is great for data science and machine learning tasks — and it seamlessly integrates with Gemini to write code for you!

To access Colab, you can type:… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI