Kangas, an open-source tool developed by Comet ML, is emerging as a powerful resource for processing and analyzing multimedia datasets in the field of computer vision, similar to the role Pandas plays for tabular data. Kangas facilitates the exploration, visualization, and manipulation of large-scale image, video, and audio datasets, providing machine learning professionals with the ability to gain insights and accelerate model development. Unlike Pandas, which is limited to structured data, Kangas supports unstructured data types and offers an interactive DataGrid to display and query data efficiently. It also integrates seamlessly with machine learning workflows, enabling users to store, share, and remotely serve data. The tool's intuitive API and capabilities make it an attractive option for data scientists working on complex computer vision projects.