Kelvin Fung
About Kelvin Fung
Kelvin Fung serves as the Lead Data, acting manager, and Solutions Engineer at Transamerica, where he has worked since 2019. He has extensive experience in data architecture and automation, particularly with large language models and cloud technologies.
Work at Transamerica
Kelvin Fung has been employed at Transamerica since 2019, where he holds the position of Lead Data, acting manager, and Solutions Engineer. In this role, he has contributed to various projects and initiatives that leverage his expertise in data management and engineering. His work is based in the Greater Denver Area, where he has been instrumental in enhancing data processes and solutions within the organization.
Education and Expertise
Kelvin Fung earned a Bachelor's degree in Computer Science from an undisclosed institution. His educational background provides a solid foundation for his technical skills in data engineering and solutions architecture. His expertise includes working with large language models, automation, and various programming frameworks, which he applies in his current and previous roles.
Background
Prior to joining Transamerica, Kelvin Fung worked at Oppenheimer & Co. Inc. from 2006 to 2012 as a Solution Architect in the SQA department. During his six years there, he gained significant experience in data architecture and engineering, which has informed his current work. His background includes leading teams and managing complex data ingestion processes.
Achievements
At Transamerica, Kelvin Fung has led the data ingestion team to architect a custom data ingestion pipeline utilizing technologies such as PySpark, Redis, and EMR. He has implemented custom logic for Airflow and RDS to achieve dynamic scaling, resulting in cost savings. Additionally, he orchestrated the transition of scheduling services to AWS Managed Airflow, enhancing development and deployment efficiency.
Technical Projects and Initiatives
Kelvin Fung has managed an open-source ChatGPT project, gaining valuable experience with large language models. He has also set up and configured AWS Bedrock to test various LLM models using different prompt engineering techniques. His technical projects include developing a customizable web-based application using Django and implementing Elastic Search backup and restore processes through Airflow DAGs.