Summary
Ziwenhan Song is a recommendation algorithms specialist with 11 years of experience, currently optimizing feed recommendation for Douyin at ByteDance. He has deep NLP and video understanding expertise from prior roles at ByteDance and Chengdu Xiaoduo, focusing on multi-turn dialogue, intent recognition and QA systems that bridge research and product needs. Trained in distributed computing and AI with dual master's degrees from University of Melbourne and ANU, he combines rigorous academic foundations with hands-on production experience. Ziwenhan is known for thinking from the user’s perspective and turning product insights into algorithmic solutions that directly improve engagement. Based in Haidian, Beijing, he brings cross-domain fluency across recommendation, dialogue systems and video understanding. His GitHub handle “dreamAdream” hints at a creative, persistent approach to engineering and experimentation.
11 years of coding experience
Australian National University
Bachelor of Engineering, Software Engineering, Bachelor of Engineering, Software Engineering at Sun Yat-Sen University
The University of Melbourne