PDFs get parsed and summarised. Images get verbatim OCR plus vision description. Video and audio get transcribed with Whisper. Everything becomes a searchable neuron — zero configuration.
Drop from anywhere on the page. Global overlay. AI analyses. Neurons appear.
PDFs, DOCX, images (PNG/JPG/WEBP/HEIC), video (MP4/MOV), audio, spreadsheets, code — each routed through the right pipeline.
Drop a whiteboard photo, screenshot, or scan. Cortex pulls the exact text and adds a vision description. Search what's in the image, not just the filename.
Meeting recordings become searchable transcripts. Keyframes get vision-described so you can jump to the moment, not scrub timelines.
Drag from Finder onto any Cortex page — the screen-wide overlay catches it. No "find the upload button" hunt. Mobile users get an explicit upload tap too.
Drag from anywhere. The overlay lights up violet across the screen.
OCR, vision, Whisper, or text extraction runs in a background worker — scoped to the active project.
Within seconds the file appears in your brain, fully searchable by content.
Still curious? Write to us.
Private beta. Limited spots. Redeem your code to jump in — or join the waitlist at the bottom of the page.