Predictive Prefetching for the Web

Hosted by Silicon Valley Data Science, ML, AI Platform

Public group


Websites are slow! Double-click shows that the average load time on a 3G network is 19 seconds! On top of that, on mobile devices, JavaScript compared to a JPG image with the same size may require x25 more processing time.

How to speed up our apps? Lazy-loading is here to help! By only loading the minimum amount of JavaScript during the initial page load we can improve the UX dramatically. However, this brings another set of questions - how to provide instant page load by mindfully prefetching the bundles, without draining the users’ mobile data plan?

In this talk, we'll see how we can create a machine learning model from a Google Analytics report. Later, by empowering static analysis techniques, we'll map this model to the lazy-loaded JavaScript chunks and apply predictive prefetching. In the last part of the presentation, we'll look at Guess.js which provides a sample implementation of these ideas.


6:20 pm - 6:30 pm Arrival and socializing
6:30 pm - 6:40 pm Opening
6:40 pm - 7:50 pm Minko Gechev, "Predictive Prefetching for the Web"
7:50 pm - 8:00 pm Q&A

About Minko Gechev:

Minko is an engineer in the Angular team at Google. He loves to transform abstract theoretical computer science concepts into robust industrial solutions. Minko’s working on tools for static code analysis and development productivity. Before joining Google he co-founded, which is was acquired by Coursera in 2019.