This work presents Automata Processing as an algorithmic paradigm in which data-processing problems are formulated as networks of non-deterministic finite automata (NFA) expressed in the Automata Network Markup Language (ANML) and its graphical counterpart ANML-G. The primary aim is to show how moving beyond conventional regular-expression usage enables automata networks—implemented as pattern-matching state-transition elements augmented with counters and related primitives—to serve as a flexible basis for solving a broad range of data-processing tasks, and to provide practical guidance for constructing such machines. The notes synthesize execution semantics and modeling techniques into a tutorial-style reference and a cookbook of worked machines, enabling readers to experiment with design patterns and compose larger solutions from reusable automata building blocks. A semiconductor implementation of a chip able to run ANML descriptions was unveiled by the Micron Corporation in 2013, providing a real-world demonstration of the practicality of the automata processing paradigm. Observations and guidance are given on the use of this semiconductor architecture for Automata Processing.