Tag: Data & AI

This article argues that the biggest shift in AI design tools isn't better mockup generation—it's the emergence of agents that work directly on the design canvas, much like coding agents now operate inside code repositories. It maps the leading agentic design tools by where their agents live (canvas-native versus code-native) and contends that the most important strategic decision isn't which tool has the best AI, but where a team's source of truth resides: the design file, the codebase or the running product. Ultimately, the piece concludes that a well-structured design system has become the foundation of effective AI-assisted design, making workflow architecture—not flashy features—the key factor in choosing the right tool.

Traditional RAG systems are designed around a simple idea: find the most relevant information, add it to the prompt and generate an answer. This works well for questions where the answer can be found in a small set of relevant documents.

Keyword search and semantic search each solve different retrieval problems, but neither is perfect on its own.

RAG changes what a model knows by retrieving from a live knowledge base, while fine-tuning changes how a model behaves by training it on curated examples. In this post, we compare them in terms of accuracy, cost and citations, and show when to reach for each.

Named Entity Recognition (NER) finds the people, organizations and dates inside unstructured text and stores them as metadata alongside the embeddings. In this post, we explain how that metadata sharpens retrieval and show how Progress Agentic RAG automatically detects entities during ingestion.