Let AI pick your video thumbnails
Set smart to true on /video/thumbs, and a vision model scores candidate frames and returns the ones that make the best previews.
Kevin van Zonneveld
Here are our posts tagged with ai
Set smart to true on /video/thumbs, and a vision model scores candidate frames and returns the ones that make the best previews.

The /document/extract Robot is now generally available. Pull text and images out of PDFs for search, RAG, and archives, in the same Assembly as the upload.

Our Node SDK now ships with Intent commands: high-level CLI commands for common tasks like Markdown conversion, image generation, background removal, and more.

Fix small, flat photos in the Assembly that already resizes and stores them. /image/enhance corrects tone, and /image/upscale adds pixels with AI.

Build a resume parsing pipeline with Transloadit’s document conversion and AI robots.

Build a production-ready invoice parser by combining Transloadit’s document conversion and AI robots.

Build a business card scanner that extracts contact details from images with Transloadit.

Build a Reducto-style document intelligence pipeline with Transloadit’s OCR, split, and AI Robots, plus flexible storage and workflow primitives.

Transloadit publishes llms.txt as a compact retrieval index for documentation tools and AI agents.

Sign up for our monthly newsletter to receive direct links to 3 exclusive tech — and 2 product updates. No less, no more.