In this video, Esther Lloyd talks with Martin Omander about the architecture her customers use for real world, AI enabled finance applications. Watch along and see how development teams combine Firebase and Google Cloud to build complete production systems faster.
The architecture addresses several critical design and operational concerns, like how to maintain a single codebase for iOS and Android apps, protect APIs and models from unauthorized access, abuse, and prompt injections, avoid AI vendor lock-in, run at scale with zero server management overhead, and monitor app crashes, latency, and model performance in a single unified view.
Chapters:
0:00 – Intro
1:19 – Our example: Cymbal Finance
2:03 – Flutter and Firebase
4:55 – Google Cloud architecture
6:39 – Takeaways
🔗 Resources:
* Architecture guide: Deploying and Operating Generative AI Applications → https://goo.gle/4grPAqi
* Google Agent Skills on GitHub → https://goo.gle/45qXiL5
* Dart/Flutter MCP server → https://goo.gle/3TTDAFm
* Firebase MCP server → https://goo.gle/4bI4kii
Watch more Serverless Expeditions → https://goo.gle/ServerlessExpeditions
🔔 Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech
#ServerlessExpeditions #GoogleCloud
Speakers: Martin Omander, Esther Lloyd
Products Mentioned: Flutter, Firebase, Firebase Authentication, Identity Platform, Firebase App Check, Firebase AI Logic SDK, Firebase Remote Config, Firebase Cloud Messaging, Firebase Performance Monitoring, Crashlytics, Google Apigee, Google Cloud Run, Model Garden, Model Armor, Gemini Enterprise Agent Platform, Cloud Operations Suite, Gemini, Anthropic Claude, Dart, Flutter MCP Server, Firebase MCP Server
