Yin Tang

Hello! I’m Yin Tang, a passionate coder with a keen interest in solving real-world problems more efficiently and creatively. My areas of interest span machine learning, computer graphics, game development, and blockchain. I believe these fields will intersect at some point in the future.

I am currently a second-year Master’s student at the University of Pennsylvania with major in Computer Graphics and Game Technology. I have hands-on experience working with OpenGL, Qt, Unity as well as Unreal Engine. I received my BA in Computer Science and Applied Mathematics from the University of California, Berkeley in 2021, where I developed a strong foundation in programming and problem-solving. You can explore my projects and view my resume to learn more about my work.

For the past summer, I interned at Plus, an industry leader in autonomous driving truck solutions, working on an internal data infrastructure tool designed to search for relevant images from truck bag data using text/image queries. My internship project focused on integrating Milvus, a vector database specialized in embedding similarity search, into the existing production architecture to reduce latency and memory usage. By using Milvus with compression index IVF_SQ8 and advanced memory mapping feature, I was able to reduce both runtime latency and memory usage by 10 times. Leveraging the freed-up resources, I deployed a new LVM, Florence-2-large with Stella, for embedding generation and runtime query, which greatly enhanced search quality by creating embeddings from image captions that offered more detailed descriptions. Additionally, I developed a proof-of-concept for video captioning using the LLaVA model and set up a testing environment for my colleagues to continue the project. I also built a prototype Human-Machine Interface (HMI) in Unreal Engine to visualize real-time driving data. I learned a great deal from working on the comprehensive infrastructure pipeline, which further sharpened my modular programming skills and enhanced my full-stack understanding.

After graduating from Berkeley, I worked as the Product Manager at Bobber, a startup in the Web3 industry. For the first six months, I applied what I had learned in school to iterate our product by constructing routine OTA packages and performing system debugging, which significantly enhanced my coding and debugging skills at the system level. Starting in the second half of my first year, I was soon given broader responsibilities. I managed a team of 3 to oversee the manufacture and maintenance of our hotspots, coordinated with downstream manufacturers through routine sync-up meetings, and co-led support engineering efforts for remote technical support. I also directed the development of 3 new products from design to prototyping, which required extensive self-learning and from which I significantly improved my project management, product development, and communication skills. Joining Bobcatminer as one of the first 10 employees was a rewarding experience that has greatly contributed to my career growth.

I am currently seeking full-time opportunities as a Software Engineer, Graphics Engineer, or Machine Learning Engineer. If you have any opportunities or would like to connect, please feel free to reach out at jackeyty@berkeley.edu!