New comment by bckmn in 'Show HN: Tarsier – Vision utilities for web interaction agents'
Reminds me of [Language as Intermediate Representation](https://chrisvoncsefalvay.com/posts/lair/) - LLMs are optimized for language, so translate an image into language and they'll do better at modeling it.
Reference
Widgets
Insight Agent
This widget generates “insights” about the post using an agentic loop and MCP server for this site. A response may take up to a minute to generate.
Generating