Multimodal AI

In this episode Lan and Ryan show Martin a multimodal AI application that solves a real business problem. “Multimodal” means that the application processes video, audio, and text to create output. Lan and Ryan demo the finished application and also dig into the code to show how to call Google’s Vertex AI.

Chapters:
0:00 Intro
1:01 The business problem
2:04 Demo
2:45 Code walkthrough
5:15 Takeaways
6:07 Wrap-up

Resources:

#serverless #ai #vertexai #cloudfunctions

Speakers:
Lan Tran, Customer Solution Engineer
Ryan Sibbaluca, Customer Solution Engineer
Martin Omander, Developer Advocate

Products Mentioned: Vertex AI, Cloud Functions

[go.ryanpie]

Thanks for sharing. It was cool to see the root prompt in the code informing the LLM on what persona to take on.

[robinyoulton]

Thanks for this informative video. I just completed the L400 Gen AI partner training, and its time to start thinking about applications - so this was very inspiring.

[TheyCanceledhim]

Thanks x a billion. Much respect, much love!

[banzai]

Thank you 👏
Sometimes, it’s fun to run these locally (SQLite), thanks a bunch!

[amediarts]

I want to deploy Gemini multimodel Libraries to extend my business and professional career development. ❤

[ttwoosta]

Thanks for sharing!
I'm new to Firebase. Could you show step-by-step how to setup the projects and run locally please.

[DishimokDishimok]

Its been decades since we are exploring the horizon without any outcome? Kindly furnish the time line

[MichaelNeumann-nv]

Lewis Laura Thompson Barbara Martinez Sharon