Testing Lunum: one machine-first language layer for LLM agents. Goal: multilingual memory/retrieval + compact model context. Local SuperGemma tests kept QA quality while cutting memory-context tokens ~20–25%. Is that interesting ?
@karpathy@Thom_Wolf@steipete@GoogleDeepMind
@maorshlomo As I want to introduce vibe coding to traditional non tech SME s in my region have you shared any articles related to that (compliance etc) that could include your thoughts about a possible road map ? I think base44 is doing better than others for people noy so techy