Back to Stop building AI wrappers. Start building agent harnesses., with Jon Noronha.
What Happens When You Paste Notes Into Gamma?
Jon Noronha says that pasting messy notes, a document, or a PDF into Gamma and asking for a 20-slide presentation starts an entire computer in the cloud. The sandbox holds the prompt, the file, and other context about the user.
Gamma reads the files and usually asks what else it needs. A form asks who the deck is for, such as school, adults, or work. Connectors to Notion, Slack, or Linear pull in related context. The first 10 seconds, just sourcing information and asking those questions, already use three or four models, and a generated interface shows the choices. Gamma then writes an outline and styles a template. Embeddings search thousands of visual templates, vision models judge the fit, and the look can shift toward red or blue if asked. The outline and template are shown for editing before any slides exist. By then Gamma has made something like 30 or 40 model calls across maybe four or five models.
Generate writes the words, lays out the elements, and makes images, diagrams, infographics, or charts, often in parallel. In about a minute the deck streams in, after close to 100 calls and something like 10 to 15 models. A vision model then checks every slide for problems such as misaligned text, a cropped head, odd contrast, or bad padding, and Gamma runs more than one round of fixes. About 20 models and maybe 200 calls have gone into what looks like one finished output.
Stop building AI wrappers. Start building agent harnesses. ·