Eradicating friction for client side AI agents with Firebase

Google Cloud Tech · Beginner ·🤖 AI Agents & Automation ·1h ago
In this interview from Google I/O, Greg Bagues sits down with Seba Gnagarella, Lead Engineer for Firebase, to explore how the platform is morphing into the ultimate client side launchpad for agentic development. Learn how a decade of well structured documentation, clean APIs, and powerful CLI tools have made Firebase the ideal playground for coding agents and full stack automation frameworks. This discussion covers major architecture updates, including the relaunch of Firebase SQL Connect with real time replication and offline support, the integration of Firebase CI Logic running Gemini 3.5 Flash, and frictionless Google Workspace OAuth mappings. For cloud engineers and system architects, this session outlines why meeting the agent where it is building, whether inside Android Studio, Google AI Studio, or Antigravity IDEs, is the next frontier in lowering transaction costs and scaling localized edge persistence securely. Watch more Google I/O Interviews → https://goo.gle/io-tech-chats 🔔 Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech #GoogleIO #GoogleCloud Speakers: Greg Bagues, Seba Gnagarella Products Mentioned: Firebase, Firebase SQL Connect, Gemini, Google Workspace, Android Studio, Google AI Studio, Antigravity
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related AI Lessons

AiFinPay: Autonomous Payments for cirosantilli/china-dictatorship
Learn to implement autonomous payments for AI agents using AiFinPay and get started with a quick installation guide
Dev.to AI
Sharing Your .env With LLMs Is Relatively Safe. Is It Really? Here’s Why.
Sharing .env files with LLMs may not be as safe as thought due to agentic attack surfaces, learn why and how to mitigate risks
Medium · Cybersecurity
From ELIZA to Agents — How AI Changed Everything and Then Changed Again
Discover how AI evolved from ELIZA to agents, transforming the field and its applications
Medium · Machine Learning
The Compliance Myth: How Modern Ecosystems Are Replacing Tick-Box Learning
Discover how modern ecosystems are replacing traditional tick-box learning with immersive, AI-driven experiences, revolutionizing corporate training and compliance.
Dev.to AI
Up next
AI Dev 26 x SF | Or Dagan: Optimizing Accuracy, Cost, and Latency in Real-World Agents
DeepLearningAI
Watch →