Gamma co-founder Jon Noronha joins Greg Isenberg to explain agent harnessesAn AI agent harness is the software framework that packages a model with tools, instructions, context management, execution controls, and user interaction. as the tools and workflows that let a model act, inspect work and incorporate user preferences. He traces how Gamma builds and visually checks a presentation using many model calls.
They discuss opportunities in specialized and subjective work, the challenge of generic AI output, and the limits of objective evaluationsEvaluation is the systematic assessment of a system's behaviour or results against explicit criteria. for personal taste. Jon Noronha describes using experiments and user behavior to compare models and product choices.
The conversation covers model selectionModel selection chooses the AI model whose capability, quality, cost, speed, safety, and operating constraints best fit a task., prompts and fine-tuningFine-tuning adapts a pretrained model by updating its parameters on examples chosen for a particular task or subject., a fast-mode experiment, marketing automationWorkflow automation uses software or AI to complete a connected sequence of routine steps with less manual effort. and image-model applications. Anthropic sponsored the episode.
Watch on YouTube
