How Bast AI uses DVC as a data registry for unstructured AI pipelines—versioning PDFs, page images, ontologies, and retrieval context to build an explainable, offline-ready medical assistant with full provenance and auditability.
This post describes a production ML pipeline for fine-tuning large language models using DVC, SkyPilot, HuggingFace Transformers, and quantization techniques.
Alex Kim
September 8, 2023
13 minutes read
We use cookies to improve your experience and understand how our site is used.