During my time in the Parallel Computing Department at Huawei, I worked on enhancing the parallelism of state-of-the-art transformer models. This involved studying transformers from a mathematical perspective, establishing a formalized mathematical framework to optimize their structure, and implementing improvements.
Collaborated with Huawei's Network Team to infer and analyze throughput data, focusing on enhancing and optimizing predictive modeling. Employed a range of Machine Learning models to develop datasets containing questions related to telecommunications, leveraging Large Language Models for improved insights.
Languages
Programming Languages