Hello! I’m Yerrapothu Venkata Kesava Siva Sai, a first-year Master’s (M1) student. I recently had the opportunity to participate in SIG-CDS43, the 43rd CDS Symposium hosted by the Consumer Devices & Systems (CDS) Study Group of the Information Processing Society of Japan. The symposium was held in Hanoi, Vietnam, from Thursday, May 29th to Friday, May 30th, 2025.
This event brought together researchers working on cutting-edge topics in consumer devices and systems, many of which aligned closely with my own field of study. I was honored to present my research titled:
“An Evaluation Method for CloudFormation Created by Generative AI.”
In my study, I explored how generative AI tools perform when tasked with producing AWS CloudFormation templates. The findings showed that AI-generated templates had a 60% success rate on My Research using LocalStack had got 75%, compared to 90% for human-written templates. The study also highlighted frequent issues like overly permissive IAM roles. However, I found that prompt tuning can significantly improve AI output, though thorough validation remains crucial.
This experience not only allowed me to share my work with a broader community but also helped me gain valuable insights and feedback for future research.
After the conference, we attended a dinner party with professors and students from different Universities in all over the Japan. Over great food and a relaxed atmosphere, we enjoyed lively conversations about research, future career opportunities, and ongoing academic projects. It was a memorable evening that helped build strong connections and left me inspired for my future work.
