jsxyzb

Currently navigating the technical hurdles of video diffusion models, long-context window optimization, and RAG efficiency. I spend most of my time debugging inference latency and fine-tuning text-to-presentation pipelines.

Passionate about bridging the gap between frontier research and production-ready tools.