GitHub user sahil-endee added a comment to the discussion: Ambient "operator recommender" — predictive next-operator suggestions on the canvas
Sensible next step: I'll open a small end-to-end PR that wires up a stateless /recommend endpoint in the existing agent-service (reusing the compile API for schemas, the operator catalog, and the same AI SDK + LiteLLM gateway the chatbot already uses), and a frontend hook that listens for "operator added" and feeds suggestions into the existing ghost-operator UI from drag-drop. First version returns hardcoded suggestions so we can confirm the wiring works; second version swaps in a real LLM call with a small/cheap model. Would that be ok? GitHub link: https://github.com/apache/texera/discussions/5240#discussioncomment-17397470 ---- This is an automatically sent email for [email protected]. To unsubscribe, please send an email to: [email protected]
