NLP
Natural language processing — from n-grams and embeddings to sequence-to-sequence and transformer models, studied in my IBM coursework.
Natural language processing (NLP) is the field of getting computers to understand and generate human language: tokenization, embeddings, language models, translation and classification.
How I use it
I covered classic and neural NLP in the IBM Generative AI Engineering Professional Certificate professional certificate — n-gram and sequence-to-sequence models in Generative AI Foundational Models for NLP & Language Understanding, NLP data loaders in Generative AI and LLMs: Architecture and Data Preparation and translation with Transformers in Generative AI Language Modeling with Transformers. In practice I apply NLP through LLMs, for example in NovaMX CRM's lead and property classification engine and in RAG (Retrieval-Augmented Generation) retrieval.
Why it matters
Those foundations make the jump from classification tasks to LLM-based pipelines much easier to reason about. Classic NLP concepts — tokenization, embeddings, sequence models — still explain much of how LLM-based systems behave, from context limits to why embedding-based retrieval works.