Generative AI offers immense power, but with it comes significant risk. This article discuss how to protect your LLM applications from prompt injection, data leakage, and other threats using a multi-layered security approach with Google Cloud services like Natural Language API, Model Armor, and vector databases.
Your AI agents are only as good as their ability to collaborate. As we move from single agents using tools to multi-agent ecosystems, our methods for evaluating success must evolve. Learn why trajectory and handoffs matter more than ever.
Discover how Media CDN's new flexible origin shielding enhances content delivery! Learn how to reduce origin load, optimize latency, and improve viewer experience by precisely controlling your shield locations.
There is a common refrain from many that “why can’t agents just be tools?”. This blog provides a point of view about why tools and agents should be treated differently, and why there needs to be a different way to interact with them.
Learn how about NVIDIA's open-sourced library for optimizing LLM inference directly into Vertex AI Prediction, enabling you to serve open models with 1-click deployment and getting significantly improved performance and cost-efficiency.
In my previous blog From Data to Metadata: Unlocking Insights with Gen AI in BigQuery, we explored how BigQuery's “Insights” feature can automatically generate metadata for your tables and columns helping to uncover hidden patterns and trends within your data. We are now taking it a step further with metadata and insights generation at the dataset level. This new capability not only describes the dataset but also provides an interactive entity relationship map that visually reveals connections between different entities.
We recently announced the general availability of two new models from the latest generation of Anthropic’s Claude model family on Vertex AI: Claude Opus 4 and Claude Sonnet 4. In this blog, we’ll guide you through building with the new Claude 4 models on Vertex AI.
Deploying Deepseek-R1-0528, a 671B parameters model, typically necessitates at least 8x H200 GPUs for a single API request. Such extensive resources are not accessible to everyone.To make Deepseek R1 more accessible to developers, we're excited to announce that DeepSeek R1 is now available as a fully managed API on Vertex AI in Preview.
Mirror mirror on the wall, What do I do with my API sprawl, Where can I see them all, How to make sure they listen to my call, And to ensure their compliance doesn’t fall. API hub is the one to rule them all!
| User | Likes Count |
|---|---|
| 11 | |
| 7 | |
| 4 | |
| 3 | |
| 2 |
