Join Ken Yesh for an interactive workshop where we’ll recreate the iconic "SeeFood" app from Silicon Valley. Moving beyond text-based AI, we will build a functional web application that uses multimodal intelligence to classify images.

This session is designed as a guided build-along, focusing on how "agent-first" tools like Google Antigravity can accelerate the move from a product concept to a working MVP.

What we’ll build together

• The Interface: Use Antigravity to scaffold a modern web UI with an image upload and preview system.

• The Intelligence: Connect your app to Gemini 2.5 Flash to perform instant, multimodal image analysis and classification.

• The Deployment: If time and prompt tokens allow, we’ll use Antigravity’s agentic features to streamline the process of pushing our project live to Google Cloud (Cloud Run).

You’ll leave the session with a functional AI application and a practical understanding of how to orchestrate autonomous agents to handle the heavy lifting of modern development.

Requirements

1. Laptop with the Antigravity application installed.

2. A Google Cloud account (free tier eligible).

3. A Gemini API Key from Google AI Studio (strongly recommended to generate this before the session).

About the Speaker

As an ML Product Manager at Apella, Ken Yesh leads the development of computer vision models for surgical analysis. He also serves as a product advisor to early-stage startups, bridging the gap between core product fundamentals and the adoption of agentic workflows. A power user of Google Antigravity, Ken focuses on the intersection of product strategy and autonomous development, regularly building tools like FGAC.ai to explore the future of agentic engineering.

Thank you to our hosts!

A big shout out to Cahoots Ann Arbor for hosting our gathering! They've been a pleasure to work with, and one of Ann Arbor's premier coworking spaces. Check them out on Instagram or LinkedIn!