Tiny Machine Learning (TinyML) is a subset of Machine Learning (ML) application deployment where design focus is placed on transitioning memory and compute intensive ML models, typically trained on the cloud in large data-centres, into resource constrained edge devices. The motivation for this transition is to enable better energy efficiency, preserve user data security and minimize network usage. This chapter explores the deployment of a TinyML application through a pipeline of algorithm exploration, co-design techniques to optimize these algorithms for edge devices, and custom and commercial implementations of edge device architectures. In particular the chapter focuses on current developments with Neural Network based approaches that offer state of the art performance and functionality as well as an emerging algorithm called the Tsetlin Machine to understand whether this recently proposed method ca be a suitable alternative. This algorithm-to-hardware approach is now developed into popular automation flows, libraries and toolchains; this chapter examines the scope, usability and drawbacks of these tools and platforms.