OpenAI Engineer Asks Simon Willison for Model Vibe Check
OpenAI engineer Vaibhav Srivastav asks Simon Willison for a fresh model evaluation method.
Vaibhav Srivastav posted a reply to Simon Willison. He noted that the pelican on a bike SVG appears saturated and asked what Simon's current go-to vibe check is for a model or harness. Srivastav is identified in the post as developer experience and community lead for EMEA at OpenAI, with prior experience at Hugging Face. The message stands as the sole item in the packet and contains no further details on responses or outcomes.
Combined views
16K
2 posts, first seen 9h ago
OpenAI Engineer Asks Simon Willison for Model Vibe Check
OpenAI engineer Vaibhav Srivastav asks Simon Willison for a fresh model evaluation method.
Vaibhav Srivastav posted a reply to Simon Willison. He noted that the pelican on a bike SVG appears saturated and asked what Simon's current go-to vibe check is for a model or harness. Srivastav is identified in the post as developer experience and community lead for EMEA at OpenAI, with prior experience at Hugging Face. The message stands as the sole item in the packet and contains no further details on responses or outcomes.
Reactions
2 postsV( 
Vaibhav (VB) Srivastav@reach_vbhey @simonw - given pelican on a bike svg is now saturated (?), what's your go-to vibe check for a model/ harness these days?
- likes: 90
- replies: 22
- bookmarks: 3
- reposts: 0
SW Simon Willison@simonw@reach_vb I'm still trying to find one! Right now I'm mostly back to ad-hoc experiments, and I still don't have a great one for long/running agent tasks yet
- likes: 37
- replies: 7
- bookmarks: 1
- reposts: 1