This project aims to research and develop technology for Dutch automated writing evaluation for adult non-academic writing. The starting point is the detection and correction of sentence-level errors such as spelling and certain grammatical errors. The current state of the art for languages such as English will be compared with the innovative new possibilites that large language models (LLMs) bring. This research will also explore the possibilities for writing evaluation and support beyond the sentence level, working on a system that can evaluate paragraph and text-level writing errors, most notably those that impact a text's overall cohesion and coherence. The final pillar of this project is the exploration of the best ways to provide automated writing feedback to people in order to motivate them to continue writing and the improve their actual writing skills in the process. To accomplish this, the use of generative language models compared to more traditional methods such as feedback templates will be thoroughly explored as well.