Here is a 2-paragraph summary for mathopen.com:
Terence Tao shares how learning to use coding agents has made it surprisingly easy to transform static research figures into interactive, animated applets. What once required significant extra effort can now be done in a straightforward way, as long as the original figure was already generated programmatically, such as with Python code.
As a concrete example, Tao converts Figure 1 from his recent paper on the Gilbreath expectation sequence into a dynamic visualization. The result gives readers a more engaging and explorable way to understand the mathematical structure behind the sequence, illustrating how modern coding tools can breathe new life into traditional research outputs.