Trust but Verify
Story Tiles™ show you the underlying source data from your actual assets used to power your answer.
Trusted by users at leading companies
Upload the files you want to work with
Begin by uploading files you want to work with, such as PDFs, presentations, or call recordings into Storytell
Storytell clusters relevant StoryChunks™ together into StoryTiles™
Once uploaded, Storytell breaks these files down into small text fragments called chunks. These chunks are organized into clusters of closely related concepts. The more closely related a concept is, the more likely it is to be included in the same cluster. Each cluster of related chunks forms a StoryTile™ representing a single idea or topic that can span across multiple files.
Storytell finds, references, and returns relevant chunks
When you ask a question or submit a prompt, Storytell dynamically generates StoryTiles™ by identifying the chunks most relevant to your query. Each StoryTiles™ is then assigned a unique identifier that’s durable and can be referenced in the future to validate the authenticity and veracity of the ground truth data from the LLM. It then uses this information to deliver the most accurate and relevant response possible.
Verify the source StoryTile™
Whenever Storytell returns a response, it also displays the source StoryTile™. This direct reference back to the exact location in the original material allows you to verify the information source and trust the accuracy of what Storytell provides you.
Stop sifting through noise & let Storytell find the signal
