Junjie Qiu

Led video-agent architecture for AI short-drama production, using narrative memory, shared references, and review feedback to guide generation and preserve continuity.

Designed and pretrained temporal models for cross-domain forecasting, learning transferable representations from variable-length history for probabilistic forecasting.

Adapted pretrained temporal models to Ascend and evaluated them against reproduced baselines on telecommunications and power forecasting datasets.

Benchmarked and commissioned HPC nodes, then used MPI profiling to optimize the interconnect topology of four newly installed nodes, reducing MPI communication time by approximately 20%.