Natural Language Processing for AI Engineers

Catalog description

This course develops students ’ ability to apply natural language processing (NLP) techniques to solve engineering problems involving human language data. Learners implement reproducible text preprocessing and tokenization pipelines, apply word embeddings, and use sequence-based and transformer-based models to complete tasks such as classification, sentiment analysis, and question answering. The course emphasizes rigorous evaluation beyond accuracy, using task-appropriate metrics, structured error analysis, and stress testing under realistic input variation, to identify failure modes and reliability risks. Learners also assess bias, safety, and responsible-use concerns (including generative AI risks when applicable) and produce governance artifacts that document intended use, limitations, evaluation results, and monitoring expectations. Through applied engineering work, learners demonstrate judgment in selecting methods and designing solutions that meet technical, ethical, and operational constraints.