Summary
Haowen Hou is an Assistant Professor and seasoned AI researcher with nine years of experience building large-scale ML systems and industry-grade multimodal models. Previously a senior AI/ML engineer at Tencent, he led e-commerce visual-language pretraining on 100M product pairs, created the BagFormer dual-encoder for cross-modal retrieval, and drove video-product identification that helped WeChat reach 500M DAU. His work blends rigorous research (PhD in ECE from NUS and research at SMART) with hands-on production skills in TensorRT, Kubernetes, FAISS and large-data engineering, delivering measurable business impact such as top competition placements and multi-percent lifts in engagement and revenue. Now focused on RWKV architectures and visual-language models at Shenzhen University, he combines academic curiosity with a proven track record of deploying scalable, real-time multimodal pipelines. An uncommon strength is his ability to translate fine-grained entity knowledge into bag-wise cross-modal interactions that improve retrieval without sacrificing latency.
9 years of coding experience
5 years of employment as a software developer
Bachelor of Engineering (B.Eng.) Engineering, Bachelor of Engineering (B.Eng.) Engineering at Harbin Institute of Technology
Doctor of Philosophy (Ph.D.) Electrical and Computer Engineering, Doctor of Philosophy (Ph.D.) Electrical and Computer Engineering at National University of Singapore
Chinese, English