How AI describes Latent Scope

Latent Scope is an open‑source tool that uses LLMs to embed, visualize, cluster, and categorize unstructured text data, combining a pipeline with an exploration interface. It can run locally with open‑source models or via third‑party providers and ingests data from CSV, Parquet, or Pandas DataFrames, producing a scoped dataset stored as parquet with JSON metadata. The pipeline performs ingest, embed, project (UMAP), cluster (HDBSCAN), and label steps, enabling exploration and curation of results in its interface.