Postcards from Next ’23 

Postcards from Next ’23 allowed attendees to create unique postcards with PaLM 2 and Imagen on Vertex AI. Guests choose different text styles and art styles at a kiosk, printed their AI-generated design, and mailed their memento to friends, family, or colleagues. Holding their postcard in hand, attendees saw how sophisticated Google Cloud’s AI capabilities are, and how they can use Vertex AI beyond Next ’23—to ideate, refine concepts, and bring written and visual worlds to life.

Postcards from Next ’23 allowed attendees to create unique postcards with PaLM 2 and Imagen on Vertex AI. Guests choose different text styles and art styles at a kiosk, printed their AI-generated design, and mailed their memento to friends, family, or colleagues. Holding their postcard in hand, attendees saw how sophisticated Google Cloud’s AI capabilities are, and how they can use Vertex AI beyond Next ’23—to ideate, refine concepts, and bring written and visual worlds to life.



Position:
Creative Technologist Intern at Deeplocal
Role:
Contributed to conceptualizing meaningful experiences for users. Explored potential iterations and tools for execution.
Tools Used: Python, Vertex AI (PaLM 2 & Imagen), Touchdesigner, Websockets

Team Lead:
Blair Neal

Client:         
Google Cloud Next’23 Conference





Role and Responsibilities:

As a Creative Technologist Intern at Deeplocal, I played a role in conceptualizing and executing this project. My responsibilities included:

  • Experience Concepting: Developed ideas for meaningful and engaging user experiences that leveraged Google’s AI technologies. This included exploring concepts such as interacting with an AI Gardener to grow virtual businesses and the Vertex AI Tree of Life to create hybrid wildflowers with unique properties.

  • Iteration and Tool Exploration: Explored various iterations and tools to bring the concept to life, ensuring a seamless and impactful user interaction. This included testing Enterprise Search's capability to handle structured and unstructured data from various sources.

  • Prototyping: Connected Vertex AI’s PaLM 2 Model via OAuth to Python and then to Touchdesigner via Websockets, allowing the interactive experience to be built inside Touchdesigner. This enabled the use of voice inputs and tangible interface inputs.



Impact and Insights:

The project effectively demonstrated the practical applications of Google Cloud’s AI technologies. This hands-on experience not only showcased the technical sophistication of Google’s AI but also provided valuable insights into how such technologies can be used to enhance user engagement and creativity in various contexts.