Welcome to my blog! Here you will find a collection of articles about recommendation systems. The articles are written by me and are based on my experience in the industry. Furthermore, I will share some knowledge that I have acquired, such as game theory and LLM. I hope you will find them useful.

In 2020, I graduated from Beihang University with a Bachelor of Engineering degree in Computer Science. Throughout my undergraduate studies, I immersed myself in a variety of computer science subjects, including operation systems, compilers, computer organization and architecture, software engineering, computer networking, etc. By designing and developing a single-cycle CPU, a C0 compiler, and complex multi-threaded software systems, I acquired practical skills and knowledge that helped me gain a holistic understanding of computer science.

In 2018, I participated in a project with the CAS studying the role of entity relationship features in text extracted by GCN for multimodal matching. The following year, I was involved in the development of a face recognition model at Megvii, where I gained exposure to vector search engines. These experiences not only provided me with a deeper understanding of natural language processing and computer vision, but they also served as a stepping stone for me to explore the field of recommendation systems.

In 2020, I had the privilege of joining Kwai as a recommendation algorithm engineer intern. During my time there, I was tasked with developing the deployment process for the recall model on the new middle platform, improving deployment efficiency and reducing operation and maintenance costs. Furthermore, I participated in the development of a new recall model that achieved significant performance in the online environment. Additionally, I collaborated with my mentor to develop and release the initial version of the Kwai general recommendation embedding, RecoEmbedding, which is now widely used in various recall and ranking models at Kwai.

In 2021, I joined ByteDance’s recommendation department. While my previous role at Kwai involved developing recall models, my main responsibility at ByteDance is to develop ranking models. The model I developed to remove position bias has delivered impressive results in the online environment. Through my participation in the development of a large-scale recommendation system, I gained invaluable knowledge and experience. This experience granted me a profound comprehension of the complexities and challenges involved in developing effective recommendation systems. It equipped me with the requisite skills and expertise to confidently address these challenges directly.

In 2022, driven by an interest in quant, I am currently pursuing my master’s degree in financial computing at the School of Informatics, University of Edinburgh.

I welcome any opportunity to engage in discussions related to the topics I address in my blog. Sharing knowledge and exchanging ideas with others is a vital part of my learning process. Therefore, I encourage you to share your thoughts and opinions with me, as I am always eager to expand my understanding and learn from others.