25-26 First T19

T19 Session is the main session of GDGoC Yonsei.

• In the APT Update Session, presenters share insights on the latest technologies and tech trends.
• In the Tech Session, presenters give talks on technologies of their own interest.

✔️ APT Update Organizer: Hyunwoo Jeon Nano Banana Nano Banana is an image editing model built on Google Gemini 2.5 Flash Image. Unlike traditional models that generate images from noise without additional context, Nano Banana leverages the full resources of Gemini, making autonomous generation possible—a clear point of differentiation. Docker Sucks Docker Sucks is an open-source platform that runs software using container technology. It isolates the elements needed to run an app, ensuring security through hardware-level separation, while also offering optimization for Apple Silicon as one of its key strengths. The Ultimate Conversational Language Model: Who Leads Today? Currently, multiple conversational language models such as ChatGPT, Gemini, Claude, and Grok are available. Comparative analysis suggests: ChatGPT stands out for general-purpose usage. Claude excels in creative writing and content generation. Gemini is advantageous for large-scale literature review and resource search. Grok shows particular strengths in mathematics and specialized technical knowledge.

✔️ Tech Session Cloud Core – Seoho Hwang [Come, come to the cloud!]

Cloud computing is the technology of renting resources such as servers, storage, and networks via the internet. As an essential tool for building services from ideas, the speaker himself was fascinated by its potential—leading him to choose the Cloud division.

ML/AI Member – Chanhee Lee [Does artificial intelligence really think like humans?]

While artificial neural networks update all connection weights at once by calculating input–output errors, the brain’s neurons dynamically adjust connections only at the local level. Recently, predictive coding models have emerged, raising expectations for AI structures that resemble how the brain operates.

DevRel Member – Taekyung Lee [Privacy issues in AI agent environments]

AI agents—systems that autonomously perform tasks on behalf of users—are becoming increasingly adopted. However, issues around personal data protection have surfaced, from LLM inference structures to leaks via external applications. As AI innovation pushes forward, the scope of data collection and the importance of development ethics must be emphasized more than ever.