HeyGen's Rui Zhang shares GPU optimization and scheduling lessons for inference costs

SGLang's X account posted on November 8, 2026, that HeyGen's Rui Zhang discusses GPU performance optimization and intelligent scheduling for lowering AI inference costs.

The post says the talk draws on lessons from running HeyGen's production AI inference and treats the GPU as the starting point for reducing serving costs.

The report is a single social media post announcing the talk. It does not include specific cost figures, benchmark results, or the individual techniques Zhang covers.