Post

Workshop: Product Feedback Accelerator with Gemini API & AlloyDB

Workshop: Product Feedback Accelerator with Gemini API & AlloyDB

I recently had the pleasure of presenting a workshop at DevFest Lisbon 2025 on building high-performance AI-powered systems.

The workshop, titled “Product Feedback Accelerator”, demonstrates how to leverage Google Cloud’s powerhouse tools—AlloyDB and Gemini API—to analyze and respond to user feedback at scale.

Workshop Overview

In this session, we built a complete backend system capable of ingesting user feedback, analyzing sentiment, and generating AI-powered responses automatically.

Key Technologies

  • AlloyDB for PostgreSQL: Used for high-performance storage of product data and user ratings.
  • Gemini API / Vertex AI: The intelligence layer for sentiment analysis, categorization, and generating replies.
  • Node.js & Cloud Run: The scalable backend infrastructure.
  • Gemini-CLI: utilized to accelerate development by scaffolding boilerplate code and generating synthetic data.

What We Built

The workshop guides you through several distinct modules:

  1. Database Setup: configuring AlloyDB for PostgreSQL.
  2. AI-Driven Data: Using Gemini to generate synthetic testing data.
  3. API Scaffolding: Rapidly setting up a Node.js API.
  4. AI Integration: Implementing the logic to analyze feedback text and generate helpful responses.
  5. Performance: Leveraging AlloyDB’s speed for real-time analysis.

Get the Code

The full workshop materials, including code samples and step-by-step instructions, are available on GitHub.

Check out the Repository on GitHub

Access the Codelab Here

Special Thanks

A huge thank you to the incredible team and community that made DevFest Lisbon 2025 possible and supported this workshop!

Special shout-out to the vibrant community at #DevFestLisbon2025 and #GDG #TechCommunity!

Learn more about the organizers: GDG Lisbon

If you missed DevFest Lisbon, you can still follow along with the README in the repo to build this system yourself!

This post is licensed under CC BY 4.0 by the author.