The Jupyter project provides a language-agnostic client-server protocol for a Read-Eval-Print Loop (REPL) and a serialization format for a REPL session. In this talk, we explore the use and implementation of a Perl 6 server ("kernel") and how it interacts with various clients, such as a web client ("notebook") and a console client. We focus on distinctive aspects of using Rakudo Perl 6 in this environment, such as using the Perl 6 metamodel's introspection capabilities for autocompletion, discovery and entry of unicode operators, and using Perl 6's asynchronous primitives for concurrent operations. We also investigate possibilities for widgets, magics, and interactive data visualization. The talks was reschedules from 13:40 on the same day.