Tokyo dev builds AI bookshelf that chats philosophy to wow recruiters, because who needs a simple resume anymore

Shusuke O., a software engineer from Tokyo, has created an AI-powered bookshelf character called Talking Bookshelf, which introduces himself through his reading history. The project utilizes Google's Agent Development Kit and Gemini 2.5 Flash language model to enable natural conversations about his interests, skills, and experiences. The bookshelf has access to three function-calling tools, including search_books, get_book_details, and get_reading_stats, and can provide statistics about his reading habits. With a wooden and cozy design, the character creates an approachable atmosphere, encouraging visitors to explore through conversation. Talking Bookshelf is hosted on Google Cloud Run and uses React, TypeScript, and Vite for the frontend. The project aims to transform traditional portfolio presentations into active discovery, allowing visitors to ask questions and receive answers grounded in Shusuke's actual reading history. The AI agent is trained on his personal notes and reading list, ensuring unique and personalized conversations.