This article discusses how generative AI is changing the design process, shifting from traditional two-way interactions to a more complex three-way model involving users, interfaces, and AI. Designers must understand the capabilities of generative models and create fluid workflows that build user trust and deliver effective outcomes.

This article explains how to define and identify agentic AI use cases, focusing on the role of agents powered by generative AI. It outlines key concepts like tasks, tools, and reasoning, and provides a framework for evaluating potential applications within organizations.

Syftr is an open-source framework designed to optimize generative AI workflows by automatically identifying Pareto-optimal configurations that balance accuracy, cost, and latency. Utilizing multi-objective Bayesian Optimization, syftr allows AI teams to efficiently explore workflow options, significantly reducing the complexity and computational cost of evaluating numerous configurations. The framework supports modular customization and integrates with various open-source libraries to enhance AI workflow design.